Loading video...

Video Failed to Load

Go Home

I've solved this. timing: - agent time: 2m54s (API Streaming + Tool Time = Agent Time) - total time: 5m12s (includes TTFT/Compactions) - able to achieve these times because it can 1 shot almost all steps until it reaches step 30. agent cost: - $2.0516 took 35 dev hours:...

brianbellx's profile picture

brian

1,295 subscribers

111,829 views • 7 months ago •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

holy sh*t, it can't be f**king real Kimi K3, Claude Opus, GPT-5.6, and Gemini 3.1 Pro are FREE on CleanAPIs right now 5 million tokens a month, no card, one API key. swap your base URL and you're running four frontier models at $0 what you get: - Kimi K3 free (2.8T params, open-source) - Claude Opus free - GPT-5.6 free - Gemini 3.1 Pro free - and 33 models total open on the platform - streaming, tool calling, vision, reasoning included how to claim (under 1 min): step 1: go to step 2: sign up email sign-up, no card, no billing setup step 3: grab your API key from the dashboard step 4: set base URL to works in any OpenAI-compatible client, Cursor, Claude Code, Codex, Kilo Code, Cline step 5: start prompting no card, no plan, just use it important: > this is a third-party proxy so don't send any sensitive data through it > 5M tokens reset monthly, it's not unlimited > paid plans start at $3.50/mo for 50M tokens if you hit the cap > these free tiers rarely stick around, grab it while it's up
0:23

holy sh*t, it can't be f**king real Kimi K3, Claude Opus, GPT-5.6, and Gemini 3.1 Pro are FREE on CleanAPIs right now 5 million tokens a month, no card, one API key. swap your base URL and you're running four frontier models at $0 what you get: - Kimi K3 free (2.8T params, open-source) - Claude Opus free - GPT-5.6 free - Gemini 3.1 Pro free - and 33 models total open on the platform - streaming, tool calling, vision, reasoning included how to claim (under 1 min): step 1: go to step 2: sign up email sign-up, no card, no billing setup step 3: grab your API key from the dashboard step 4: set base URL to works in any OpenAI-compatible client, Cursor, Claude Code, Codex, Kilo Code, Cline step 5: start prompting no card, no plan, just use it important: > this is a third-party proxy so don't send any sensitive data through it > 5M tokens reset monthly, it's not unlimited > paid plans start at $3.50/mo for 50M tokens if you hit the cap > these free tiers rarely stick around, grab it while it's up

Rivon

29,925 views • 7 days ago

AgentRouter is handing out $125 in FREE API credits for OPUS 4.8 > No card. > GitHub sign-in. > One API key for Opus 4.8 / Opus 4.7 / Sonnet 4.6 / GLM 5.1 The useful part is not “another model gateway.” It is that you can point coding tools and agents at one OpenAI-compatible base URL instead of juggling separate keys for Claude and GLM, and others. What you get: • $125 signup balance • one API key • OpenAI-compatible endpoints • chat, responses, messages, embeddings, images, audio, rerank endpoints listed • claimed support for Cursor, Claude Code, Hermes, and other agent tools The flow is simple: 1. go to AgentRouter (link in comment) 2. sign in or create an account 3. generate an API key in the console 4. copy the base URL 5. paste it into Cursor / Claude Code / your agent runtime 6. run a small test before moving anything serious Caveat: AgentRouter’s own site currently has a notice saying Claude-series service was recently hit by stability issues and large-scale Claude access was suspended. So don’t treat this as “guaranteed free Claude forever.” Treat it as $125 of free routing credits to test which models are live, how the latency feels, and whether the gateway is stable enough for your workflow. Still a very solid save if you build with agents. Most people burn paid API credits just testing configs. This gives you room to test first, then decide if it belongs in your stack.
0:22

AgentRouter is handing out $125 in FREE API credits for OPUS 4.8 > No card. > GitHub sign-in. > One API key for Opus 4.8 / Opus 4.7 / Sonnet 4.6 / GLM 5.1 The useful part is not “another model gateway.” It is that you can point coding tools and agents at one OpenAI-compatible base URL instead of juggling separate keys for Claude and GLM, and others. What you get: • $125 signup balance • one API key • OpenAI-compatible endpoints • chat, responses, messages, embeddings, images, audio, rerank endpoints listed • claimed support for Cursor, Claude Code, Hermes, and other agent tools The flow is simple: 1. go to AgentRouter (link in comment) 2. sign in or create an account 3. generate an API key in the console 4. copy the base URL 5. paste it into Cursor / Claude Code / your agent runtime 6. run a small test before moving anything serious Caveat: AgentRouter’s own site currently has a notice saying Claude-series service was recently hit by stability issues and large-scale Claude access was suspended. So don’t treat this as “guaranteed free Claude forever.” Treat it as $125 of free routing credits to test which models are live, how the latency feels, and whether the gateway is stable enough for your workflow. Still a very solid save if you build with agents. Most people burn paid API credits just testing configs. This gives you room to test first, then decide if it belongs in your stack.

kocer

25,316 views • 3 months ago

I just got $20 in free credits for Claude, GPT-5.5, Veo 3, Seedance and ElevenLabs. All in one workspace. No separate subscriptions. Promo code inside. > The platform is QuadcodeAI. One workspace that replaces every AI tool you're paying for separately right now. Text, code, images, video, voice, 3D - all in one place. Claude Opus, GPT-5.5, Gemini, Grok, DeepSeek, Veo 3, Seedance, Flux 2 Pro, ElevenLabs, Suno and 40+ other frontier models. No API keys. Here's how to get $20 free: → Go to quadcodeai(.)com → Register → Enter promo code QCWELCOME0906 → 1,000 credits land instantly - worth $20 on the paid plan >> With these credits I turned my photo into a live GTA 6 character in one prompt. The Vibe Coding part is what makes this different. You describe what you want to build in plain English - the agent generates the full UI environment, builds the workflow, connects the tools, and delivers a working result. Character pipeline, product ad, app UI - the agent handles the entire chain internally without you touching a single external API. I tested it myself - fully working. Bookmark this before the promo code expires.
0:40

I just got $20 in free credits for Claude, GPT-5.5, Veo 3, Seedance and ElevenLabs. All in one workspace. No separate subscriptions. Promo code inside. > The platform is QuadcodeAI. One workspace that replaces every AI tool you're paying for separately right now. Text, code, images, video, voice, 3D - all in one place. Claude Opus, GPT-5.5, Gemini, Grok, DeepSeek, Veo 3, Seedance, Flux 2 Pro, ElevenLabs, Suno and 40+ other frontier models. No API keys. Here's how to get $20 free: → Go to quadcodeai(.)com → Register → Enter promo code QCWELCOME0906 → 1,000 credits land instantly - worth $20 on the paid plan >> With these credits I turned my photo into a live GTA 6 character in one prompt. The Vibe Coding part is what makes this different. You describe what you want to build in plain English - the agent generates the full UI environment, builds the workflow, connects the tools, and delivers a working result. Character pipeline, product ad, app UI - the agent handles the entire chain internally without you touching a single external API. I tested it myself - fully working. Bookmark this before the promo code expires.

Atenov int.

14,269 views • 3 months ago

RIP OpenClaw. How to use Claude Opus 4.6 + n8n to create a secure, autonomous agent available on all your devices: Step 1: Install Desktop Commander (Docker) Step 2: Configure permissions (mounted folders) Step 3: Create a secure connection Works today: - Run mcp-proxy - Set up a Cloudflare tunnel - Use a custom domain (from $5/year) An alternative coming soon: Use Desktop Commander Remote (free, Beta) Step 4: Create an n8n agent with Opus 4.6: - n8n VPS ($4.99/mo) - Add a Telegram trigger - Add Desktop Commander MCP - Plug integrations (Gmail, Drive, Notion, Stripe) - Add memory, subagents, and the Ralph Wiggum loop - Consider one extra sandbox (Docker, the same VPS) Done. In this setup your agent: - Can't access your API keys - Can't modify its environment - Can't access folders you haven't shared - Can't access tools you haven't approved - Must get your confirmation, e.g., when sending emails Unlike in OpenClaw, those are hard guardrails, not suggestions anyone can hack. But it can still: - Reply to your Telegram or Slack messages - Access selected folders from your laptop - Access Gmail, Drive, Notion, Linear, etc. - Install new local tools in a sandbox - Run autonomously for hours - Create multiple subagents - Learn from experience - Wake up regularly Want a detailed guide? I will break everything down + setup instructions + n8n templates on Monday. Follow + 🔔to get notified: Paweł Huryn
0:30

RIP OpenClaw. How to use Claude Opus 4.6 + n8n to create a secure, autonomous agent available on all your devices: Step 1: Install Desktop Commander (Docker) Step 2: Configure permissions (mounted folders) Step 3: Create a secure connection Works today: - Run mcp-proxy - Set up a Cloudflare tunnel - Use a custom domain (from $5/year) An alternative coming soon: Use Desktop Commander Remote (free, Beta) Step 4: Create an n8n agent with Opus 4.6: - n8n VPS ($4.99/mo) - Add a Telegram trigger - Add Desktop Commander MCP - Plug integrations (Gmail, Drive, Notion, Stripe) - Add memory, subagents, and the Ralph Wiggum loop - Consider one extra sandbox (Docker, the same VPS) Done. In this setup your agent: - Can't access your API keys - Can't modify its environment - Can't access folders you haven't shared - Can't access tools you haven't approved - Must get your confirmation, e.g., when sending emails Unlike in OpenClaw, those are hard guardrails, not suggestions anyone can hack. But it can still: - Reply to your Telegram or Slack messages - Access selected folders from your laptop - Access Gmail, Drive, Notion, Linear, etc. - Install new local tools in a sandbox - Run autonomously for hours - Create multiple subagents - Learn from experience - Wake up regularly Want a detailed guide? I will break everything down + setup instructions + n8n templates on Monday. Follow + 🔔to get notified: Paweł Huryn

Paweł Huryn

210,827 views • 7 months ago

I just compared Claude Code vs Codex vs Cursor CLI The task was to build a Next.js app with Tailwind 4 and shadcn components to collect customer feedback and showcase it with a widget. I gave all three the same prompt and let them go for 30 minutes to see what they came up with. Claude Code with Opus 4.1 Even though I told it to set up the app in the existing project folder, it tried to create a directory for it. After I interrupted and told it not to do that, it built a demo form and landing page with no errors. I had to ask it to make the demo interactive so users could submit a testimonial and preview it. The landing page looked like AI and was pretty basic, but it worked and it was done in a fraction of the time of the others. Total tokens used: 33k Codex with GPT-5 At the end of the 30 minutes I just could not get Codex to produce a working app. It got stuck in a loop of not being able to set up Tailwind 4 and despite many, MANY, attempts, I ended up with a "failed to compile" error. Total tokens used: 102k Cursor Agent with GPT-5 This was the slowest agent by far and a couple of times I actually thought it got stuck in a loop and was close to Ctrl+C'ing to cancel it. The TUI is really nice though, especially how it shows diffs and it did eventually build a working app (after one or two slight errors that needed fixing) The demo was interactive and it had a very minimal design that looked bare but also a lot less like an "AI generated" app than the Opus 4.1 design. It also wasn't too chatty and just did what it needed to do! Code quality was on a par with Opus 4.1, but it did use 5.5x as many tokens to get there. Still cheaper than Opus on a direct comparison but not when you factor in a Claude Code Max subscription. Total tokens: 188k I'll be able to do a proper comparison and record some videos when I'm back from holiday but for now, Opus is still the more capable model out of the box and Claude Code is the more complete CLI product. It will be interesting to see how Cursor evolve their CLI though with commands and subagents because I think with GPT-5 they have a real shot at providing competition for Claude Code if they can optimise output to get similar quality with less tokens. Jump to 0:40 in the video to see the two apps. Which do you think is which? ;)
1:18

I just compared Claude Code vs Codex vs Cursor CLI The task was to build a Next.js app with Tailwind 4 and shadcn components to collect customer feedback and showcase it with a widget. I gave all three the same prompt and let them go for 30 minutes to see what they came up with. Claude Code with Opus 4.1 Even though I told it to set up the app in the existing project folder, it tried to create a directory for it. After I interrupted and told it not to do that, it built a demo form and landing page with no errors. I had to ask it to make the demo interactive so users could submit a testimonial and preview it. The landing page looked like AI and was pretty basic, but it worked and it was done in a fraction of the time of the others. Total tokens used: 33k Codex with GPT-5 At the end of the 30 minutes I just could not get Codex to produce a working app. It got stuck in a loop of not being able to set up Tailwind 4 and despite many, MANY, attempts, I ended up with a "failed to compile" error. Total tokens used: 102k Cursor Agent with GPT-5 This was the slowest agent by far and a couple of times I actually thought it got stuck in a loop and was close to Ctrl+C'ing to cancel it. The TUI is really nice though, especially how it shows diffs and it did eventually build a working app (after one or two slight errors that needed fixing) The demo was interactive and it had a very minimal design that looked bare but also a lot less like an "AI generated" app than the Opus 4.1 design. It also wasn't too chatty and just did what it needed to do! Code quality was on a par with Opus 4.1, but it did use 5.5x as many tokens to get there. Still cheaper than Opus on a direct comparison but not when you factor in a Claude Code Max subscription. Total tokens: 188k I'll be able to do a proper comparison and record some videos when I'm back from holiday but for now, Opus is still the more capable model out of the box and Claude Code is the more complete CLI product. It will be interesting to see how Cursor evolve their CLI though with commands and subagents because I think with GPT-5 they have a real shot at providing competition for Claude Code if they can optimise output to get similar quality with less tokens. Jump to 0:40 in the video to see the two apps. Which do you think is which? ;)

Ian Nuttall

195,173 views • 1 year ago

The latest RAG trend for the current agent harnesses (Codex, Cowork) is to do two passes of document processing to solve a knowledge work task over a data room of documents: 1️⃣ A fast and light pass, oftentimes using a free/OSS doc parsing tool. This can be cheaply run across 10-100-1k’s of files, and enables the agent to then do retrieval (e.g. grep, semantic) to find relevant subsets of context. 2️⃣ A “just-in-time” VLM-based pass. Once the agent finds the relevant pages of context, it will screenshot the documents can call its own VLM (or write code) to dissect the pages. The issue with only using VLM-based OCR tools over massive ad-hoc customer file dumps is that it’s slow and expensive. Doing JIT VLM OCR allows the agent to filter through the data cheaply, but still preserve accuracy for the context that’s needed for the task. The agent harnesses do two-pass document processing by default using off the shelf-tools: pdf2text as the first pass, and using itself (Opus 5) as the second pass. See the below video where Cowork runs over a bunch of PDFs to answer a question about a benchmark graph in the Kimi k3 paper. The main issues here with the “out of the box” doc processing these agents offer are: * Opus 5 is not the best VLM for OCR. It is also way too expensive at scale and lacks grounding * The OSS tools like pypdf, pdf2text, may not be versatile enough as the first pass. * The agent will write a lot of throwaway code to rewrite things an OCR tool would’ve provided out of the box, like chart processing, bounding boxes, confidence scores, leading to increased cost and speed. We have all the tools within LlamaIndex 🦙 to help any agent do two-pass document processing with higher accuracy and lower cost. 1️⃣ We have liteparse for the first pass - a free/OSS parser written in Rust that’s faster/more accurate than other OSS parsers, and supports 50+ document types 2️⃣ We have LlamaParse for the second pass - an agentic document engine that uses VLMs+harnesses to achieve SOTA in accuracy and cost across various doc parsing and extraction tasks. It can be called from any agent harness as an MCP or skill. It takes in page numbers as input, so that the agent can choose to run LlamaParse over a subset of the doc instead of the full doc as a “zoom-in” pass. Come check it out! LiteParse: LlamaParse: All the relevant docs, including MCP, are here:
1:24

The latest RAG trend for the current agent harnesses (Codex, Cowork) is to do two passes of document processing to solve a knowledge work task over a data room of documents: 1️⃣ A fast and light pass, oftentimes using a free/OSS doc parsing tool. This can be cheaply run across 10-100-1k’s of files, and enables the agent to then do retrieval (e.g. grep, semantic) to find relevant subsets of context. 2️⃣ A “just-in-time” VLM-based pass. Once the agent finds the relevant pages of context, it will screenshot the documents can call its own VLM (or write code) to dissect the pages. The issue with only using VLM-based OCR tools over massive ad-hoc customer file dumps is that it’s slow and expensive. Doing JIT VLM OCR allows the agent to filter through the data cheaply, but still preserve accuracy for the context that’s needed for the task. The agent harnesses do two-pass document processing by default using off the shelf-tools: pdf2text as the first pass, and using itself (Opus 5) as the second pass. See the below video where Cowork runs over a bunch of PDFs to answer a question about a benchmark graph in the Kimi k3 paper. The main issues here with the “out of the box” doc processing these agents offer are: * Opus 5 is not the best VLM for OCR. It is also way too expensive at scale and lacks grounding * The OSS tools like pypdf, pdf2text, may not be versatile enough as the first pass. * The agent will write a lot of throwaway code to rewrite things an OCR tool would’ve provided out of the box, like chart processing, bounding boxes, confidence scores, leading to increased cost and speed. We have all the tools within LlamaIndex 🦙 to help any agent do two-pass document processing with higher accuracy and lower cost. 1️⃣ We have liteparse for the first pass - a free/OSS parser written in Rust that’s faster/more accurate than other OSS parsers, and supports 50+ document types 2️⃣ We have LlamaParse for the second pass - an agentic document engine that uses VLMs+harnesses to achieve SOTA in accuracy and cost across various doc parsing and extraction tasks. It can be called from any agent harness as an MCP or skill. It takes in page numbers as input, so that the agent can choose to run LlamaParse over a subset of the doc instead of the full doc as a “zoom-in” pass. Come check it out! LiteParse: LlamaParse: All the relevant docs, including MCP, are here:

Jerry Liu

23,119 views • 1 month ago

The same kinds of productivity gains we've seen in coding with AI agents are heading to the rest of knowledge work. This is the jump when you go from having a chatbot to being able to actually have an agent go off and do work for minutes or even hours and come back with a complete work output that you then review. Here's an example of the new Box Agent filling out an RFP response from an existing knowledge base. This process would normally take hours to fill out, and requires the full attention of the user doing the work. Now, you provide the Box Agent with the RFP questions, and it will go off, make a plan, extract all the relevant questions, read through existing source material to come up with an answer, and then generate a new word document as the final output. All while you're doing something else. The key to this architecture is that the agent is able to use all of the same tools in the background that a user uses to get work done. The agent can search for documents, read entire files, run scripts and tools in the background, and even be able to write code on the fly to automate tasks it hasn't seen before. And best of all, the Box Agent will (soon) work from the Box MCP and CLI so you can invoke it in any agentic system as a step in a process. This kind of agent complexity would have been impossible even 6 months ago. Models consistently failed at tracking long running tasks or using the right tools at the right moment for the task. But this is all now possible because of models like GPT-5.4, Opus 4.6, and Gemini 3, and is only getting better by the month. Just as we moved from engineers writing code and using AI as an assistant to answer questions, in many areas of knowledge work -like legal, finance, consulting, sales, marketing, and more- when we have a problem we'll just kick off the AI agent to just go work on it for us in the background.
2:14

The same kinds of productivity gains we've seen in coding with AI agents are heading to the rest of knowledge work. This is the jump when you go from having a chatbot to being able to actually have an agent go off and do work for minutes or even hours and come back with a complete work output that you then review. Here's an example of the new Box Agent filling out an RFP response from an existing knowledge base. This process would normally take hours to fill out, and requires the full attention of the user doing the work. Now, you provide the Box Agent with the RFP questions, and it will go off, make a plan, extract all the relevant questions, read through existing source material to come up with an answer, and then generate a new word document as the final output. All while you're doing something else. The key to this architecture is that the agent is able to use all of the same tools in the background that a user uses to get work done. The agent can search for documents, read entire files, run scripts and tools in the background, and even be able to write code on the fly to automate tasks it hasn't seen before. And best of all, the Box Agent will (soon) work from the Box MCP and CLI so you can invoke it in any agentic system as a step in a process. This kind of agent complexity would have been impossible even 6 months ago. Models consistently failed at tracking long running tasks or using the right tools at the right moment for the task. But this is all now possible because of models like GPT-5.4, Opus 4.6, and Gemini 3, and is only getting better by the month. Just as we moved from engineers writing code and using AI as an assistant to answer questions, in many areas of knowledge work -like legal, finance, consulting, sales, marketing, and more- when we have a problem we'll just kick off the AI agent to just go work on it for us in the background.

Aaron Levie

24,728 views • 5 months ago

Right now, I have a few dozen tools to automate my life. Every single thing that I do more than once every week is now a tool: either a Claude Code skill, a scheduled workflow, or an application. • I spend the time doing this once. • I schedule it. • I forget about it (or at least, try to) CREAO is one of the platforms I've used extensively for building some of these automations. They are partnering with me on this post. They are one of the only platforms where you can go from a conversation to a scheduled agent that quickly. This is how it works: 1. You describe the problem you want to solve 2. CREAO's agent builds the logic and executes it 3. You can iterate with the agent to improve the solution 4. When you are done, you turn that solution into a mini app 5. You can schedule that app to run any time you need it Something important: These agents are deterministic. They don't use an LLM to generate answers, so they will always return the same output given the same input. This is critical: when you're automating something that runs every Monday at 9 am, you need it to return the same result every time, not a "creative" answer.
5:00

Right now, I have a few dozen tools to automate my life. Every single thing that I do more than once every week is now a tool: either a Claude Code skill, a scheduled workflow, or an application. • I spend the time doing this once. • I schedule it. • I forget about it (or at least, try to) CREAO is one of the platforms I've used extensively for building some of these automations. They are partnering with me on this post. They are one of the only platforms where you can go from a conversation to a scheduled agent that quickly. This is how it works: 1. You describe the problem you want to solve 2. CREAO's agent builds the logic and executes it 3. You can iterate with the agent to improve the solution 4. When you are done, you turn that solution into a mini app 5. You can schedule that app to run any time you need it Something important: These agents are deterministic. They don't use an LLM to generate answers, so they will always return the same output given the same input. This is critical: when you're automating something that runs every Monday at 9 am, you need it to return the same result every time, not a "creative" answer.

Santiago

13,727 views • 6 months ago

HERMES AGENT LEARNS FROM ITS OWN MISTAKES. UPDATES ITS MEMORY. CREATES ITS OWN SKILLS. NO CLOUD. EVERYTHING STORED LOCALLY. THIS IS HOW THE SELF-IMPROVING LOOP WORKS. most agents start from zero every session. Hermes carries forward what it learned. THREE MEMORY SYSTEMS: 1. PROCEDURAL MEMORY (how to act) stored in ~/.hermes/skills/ as SKILL.md files. when the agent repeats a complex workflow, it saves the procedure as a reusable skill. next time the same task comes up, it follows the skill instead of figuring it out again. you can also create skills explicitly: "create a skill called video-prep that captures how I format my video scripts. spoken english, define jargon inline, no em-dashes, close with a catchphrase." the agent writes the SKILL.md. available as a slash command from that moment. Hermes ships with 90+ skills. the number grows the longer you use it. 2. SEMANTIC MEMORY (durable facts about you) stored in ~/.hermes/memory/memory.md the agent scans conversations for facts worth remembering. preferences, habits, corrections, project details. real example from the video: agent tried to scrape a YouTube channel. URL was wrong. it failed. it updated memory.md with the correct URL pattern so it never makes the same mistake again. you can also save explicitly: "save to memory that my favorite testing framework is pytest" the agent updates memory.md immediately. this file loads into context on every session. the agent knows you better every week. 3. EPISODIC MEMORY (chat history) stored in ~/.hermes/state.db (local SQLite). every conversation. every tool call. every result. searchable with FTS5 full-text search. "search our past sessions. what was the first thing I ever said to you?" the agent queries state.db and finds it. over time, auxiliary models consolidate episodic memory into semantic memory. distilling recurring patterns into durable facts. THE SELF-IMPROVING LOOP: every agent run follows this cycle: → you send a prompt → working memory loads: SOUL.md + memory.md + relevant skills + chat history → agent calls tools (terminal, browser, delegate_task) → agent completes the task, replies to you → AFTER the reply: agent checks "did I learn something worth saving?" → if yes: updates memory.md or creates a new skill → next session starts smarter than the last this happens automatically. you don't ask the agent to learn. it decides what to remember on its own. WHAT MAKES THIS DIFFERENT FROM CLAUDE CODE: Claude Code has memory too. but Hermes stores everything locally. no cloud. your data never leaves your machine. Claude Code doesn't auto-create skills from experience. Hermes turns repeated workflows into reusable procedures. Claude Code memory is instruction-based. Hermes memory is conversational and self-updating. over months of usage, Hermes builds a knowledge base of your preferences, your projects, your mistakes, and the procedures that work for your specific workflow. the agent that remembers your birthday also remembers why your last deploy failed. NO EMBEDDINGS. PLAIN TEXT. Hermes does not use embeddings or RAG for memory. skill and memory search runs on plain text keyword matching. simpler. faster. no vector database to maintain. works entirely offline on your local machine. DELEGATE TO CLAUDE CODE: Hermes can spawn a sub-agent that runs Claude Code in headless mode: "spawn a sub-agent using Claude CLI to build a Python script that fetches the top 5 Hacker News stories to markdown." Hermes delegates. Claude Code writes the code. result returns to Hermes. Hermes runs the script and delivers the output. use Hermes for orchestration. use Claude Code for heavy coding. both tools. not competitors. WHAT HERMES DOES NOT HAVE: no built-in eval or LMOps system. no LangSmith, no LangFuse integration out of the box. trajectory export and logs exist but there is no automated quality tracking. if you need eval, build it yourself or connect external tools. the loop is self-improving. measuring how well it improves is on you. comment LOOP and I'll send you the configs that control how fast Hermes learns and what it remembers. memory limits, skill auto-creation triggers, and the auxiliary model that runs the learning. Replace your entire team with 8 hermes agents👇
20:54

HERMES AGENT LEARNS FROM ITS OWN MISTAKES. UPDATES ITS MEMORY. CREATES ITS OWN SKILLS. NO CLOUD. EVERYTHING STORED LOCALLY. THIS IS HOW THE SELF-IMPROVING LOOP WORKS. most agents start from zero every session. Hermes carries forward what it learned. THREE MEMORY SYSTEMS: 1. PROCEDURAL MEMORY (how to act) stored in ~/.hermes/skills/ as SKILL.md files. when the agent repeats a complex workflow, it saves the procedure as a reusable skill. next time the same task comes up, it follows the skill instead of figuring it out again. you can also create skills explicitly: "create a skill called video-prep that captures how I format my video scripts. spoken english, define jargon inline, no em-dashes, close with a catchphrase." the agent writes the SKILL.md. available as a slash command from that moment. Hermes ships with 90+ skills. the number grows the longer you use it. 2. SEMANTIC MEMORY (durable facts about you) stored in ~/.hermes/memory/memory.md the agent scans conversations for facts worth remembering. preferences, habits, corrections, project details. real example from the video: agent tried to scrape a YouTube channel. URL was wrong. it failed. it updated memory.md with the correct URL pattern so it never makes the same mistake again. you can also save explicitly: "save to memory that my favorite testing framework is pytest" the agent updates memory.md immediately. this file loads into context on every session. the agent knows you better every week. 3. EPISODIC MEMORY (chat history) stored in ~/.hermes/state.db (local SQLite). every conversation. every tool call. every result. searchable with FTS5 full-text search. "search our past sessions. what was the first thing I ever said to you?" the agent queries state.db and finds it. over time, auxiliary models consolidate episodic memory into semantic memory. distilling recurring patterns into durable facts. THE SELF-IMPROVING LOOP: every agent run follows this cycle: → you send a prompt → working memory loads: SOUL.md + memory.md + relevant skills + chat history → agent calls tools (terminal, browser, delegate_task) → agent completes the task, replies to you → AFTER the reply: agent checks "did I learn something worth saving?" → if yes: updates memory.md or creates a new skill → next session starts smarter than the last this happens automatically. you don't ask the agent to learn. it decides what to remember on its own. WHAT MAKES THIS DIFFERENT FROM CLAUDE CODE: Claude Code has memory too. but Hermes stores everything locally. no cloud. your data never leaves your machine. Claude Code doesn't auto-create skills from experience. Hermes turns repeated workflows into reusable procedures. Claude Code memory is instruction-based. Hermes memory is conversational and self-updating. over months of usage, Hermes builds a knowledge base of your preferences, your projects, your mistakes, and the procedures that work for your specific workflow. the agent that remembers your birthday also remembers why your last deploy failed. NO EMBEDDINGS. PLAIN TEXT. Hermes does not use embeddings or RAG for memory. skill and memory search runs on plain text keyword matching. simpler. faster. no vector database to maintain. works entirely offline on your local machine. DELEGATE TO CLAUDE CODE: Hermes can spawn a sub-agent that runs Claude Code in headless mode: "spawn a sub-agent using Claude CLI to build a Python script that fetches the top 5 Hacker News stories to markdown." Hermes delegates. Claude Code writes the code. result returns to Hermes. Hermes runs the script and delivers the output. use Hermes for orchestration. use Claude Code for heavy coding. both tools. not competitors. WHAT HERMES DOES NOT HAVE: no built-in eval or LMOps system. no LangSmith, no LangFuse integration out of the box. trajectory export and logs exist but there is no automated quality tracking. if you need eval, build it yourself or connect external tools. the loop is self-improving. measuring how well it improves is on you. comment LOOP and I'll send you the configs that control how fast Hermes learns and what it remembers. memory limits, skill auto-creation triggers, and the auxiliary model that runs the learning. Replace your entire team with 8 hermes agents👇

YanXbt

22,720 views • 2 months ago

My college roommate works at OpenAI. Hasn't talked to me in 2 years. Yesterday he called out of nowhere. "Are you still doing that Polymarket thing?" I told him I run 8 Claude agents. He went quiet. "We tried building that with GPT. It doesn't work" Their agent keeps overholding losers. Exits too late. Every time. "Then Anthropic dropped Opus 4.7 and I tested it myself. Off the clock" He screen-shared. Not GPT. Claude Opus 4.7. An OpenAI researcher. Running Anthropic's model. On a $5 VPS. +$41,000 in 36 days. "I've been at OpenAI 3 years. The best agent I've ever seen runs on a competitor's model" He showed me a spreadsheet. 47 wallets. 86 million trades. Ranked by exit quality. "Top wallets capture 86% of the move and cut at 12%. Everyone else holds past 40%. GPT can't do that. 4.7 does it natively" He benchmarked Claude vs GPT on all three. Claude won every time. I asked why an OpenAI researcher is telling me to use Claude. "Because this bot makes more than my salary" My results since the switch: +$18,300. 263 trades. 80% win rate. Sharpe 2.91. Crypto +$10,800 Politics +$6,800 Weather +$4,900 Macro +$3,800. Bot for those who don't build: He said if OpenAI finds out he's running Claude bots he's done. I promised I wouldn't say his name.
0:27

My college roommate works at OpenAI. Hasn't talked to me in 2 years. Yesterday he called out of nowhere. "Are you still doing that Polymarket thing?" I told him I run 8 Claude agents. He went quiet. "We tried building that with GPT. It doesn't work" Their agent keeps overholding losers. Exits too late. Every time. "Then Anthropic dropped Opus 4.7 and I tested it myself. Off the clock" He screen-shared. Not GPT. Claude Opus 4.7. An OpenAI researcher. Running Anthropic's model. On a $5 VPS. +$41,000 in 36 days. "I've been at OpenAI 3 years. The best agent I've ever seen runs on a competitor's model" He showed me a spreadsheet. 47 wallets. 86 million trades. Ranked by exit quality. "Top wallets capture 86% of the move and cut at 12%. Everyone else holds past 40%. GPT can't do that. 4.7 does it natively" He benchmarked Claude vs GPT on all three. Claude won every time. I asked why an OpenAI researcher is telling me to use Claude. "Because this bot makes more than my salary" My results since the switch: +$18,300. 263 trades. 80% win rate. Sharpe 2.91. Crypto +$10,800 Politics +$6,800 Weather +$4,900 Macro +$3,800. Bot for those who don't build: He said if OpenAI finds out he's running Claude bots he's done. I promised I wouldn't say his name.

rari

138,929 views • 5 months ago

Jev + Opus 5.5: Anthropic's new model beats GPT-6 Astra for 1/5 the cost, and 4 API changes will 400 your agent before it writes a single line I pulled these 10 steps from the migration docs so you don't learn them in production step 1 → $4 / $20 per 1M. Opus 5 was $5 / $25. cache reads dropped from $0.50 to $0.20 step 2 → 66.4% on Terminal-Bench 4.0 vs GPT-6 Astra 57.9% and Opus 5 52.3%. +14.1 points in one release, and on FrontierCode it beats Astra at default effort for 1/5 the cost step 3 → thinking can't be turned off anymore. send thinking: disabled and you get a 400. drop the field, set effort step 4 → tool_choice any and tool are gone. 400. switch to auto + strict step 5 → edit anything above a thinking block and the request dies. append only, or opt into drop_block step 6 → computer_20251124 is dead on the API. 400. move to computer_toolset_20260801 step 7 → the quiet one: default effort fell from high to medium. your agent thinks less than you set it up to and nothing tells you step 8 → hop Opus 5.5 → Sonnet 5 → Opus 5.5 and you pay 4.36 instead of 3.32. +31%, the cache dies and Sonnet can't read Opus's reasoning step 9 → change effort at the top of the request and the cache is gone. Jev sets it per message and the cache stays step 10 → switch fast - standard mid-session and it's a full cache miss. Jev picks speed once, on turn one one model, three knobs, zero 400s. that is Jev + Opus 5.5 send this to your Claude Code before you touch the model ID, then read my full Jev deep dive in the article below ↓
0:40

Jev + Opus 5.5: Anthropic's new model beats GPT-6 Astra for 1/5 the cost, and 4 API changes will 400 your agent before it writes a single line I pulled these 10 steps from the migration docs so you don't learn them in production step 1 → $4 / $20 per 1M. Opus 5 was $5 / $25. cache reads dropped from $0.50 to $0.20 step 2 → 66.4% on Terminal-Bench 4.0 vs GPT-6 Astra 57.9% and Opus 5 52.3%. +14.1 points in one release, and on FrontierCode it beats Astra at default effort for 1/5 the cost step 3 → thinking can't be turned off anymore. send thinking: disabled and you get a 400. drop the field, set effort step 4 → tool_choice any and tool are gone. 400. switch to auto + strict step 5 → edit anything above a thinking block and the request dies. append only, or opt into drop_block step 6 → computer_20251124 is dead on the API. 400. move to computer_toolset_20260801 step 7 → the quiet one: default effort fell from high to medium. your agent thinks less than you set it up to and nothing tells you step 8 → hop Opus 5.5 → Sonnet 5 → Opus 5.5 and you pay 4.36 instead of 3.32. +31%, the cache dies and Sonnet can't read Opus's reasoning step 9 → change effort at the top of the request and the cache is gone. Jev sets it per message and the cache stays step 10 → switch fast - standard mid-session and it's a full cache miss. Jev picks speed once, on turn one one model, three knobs, zero 400s. that is Jev + Opus 5.5 send this to your Claude Code before you touch the model ID, then read my full Jev deep dive in the article below ↓

Carnage

16,674 views • 10 days ago

AI AGENTS 101 (58 minute free masterclass) send this to anyone who wants to understand ai agents, claude skills, md files, how to get the most out of AI etc in plain english: 1. chat vs agents - chat models answer questions in a back and forth while agents take a goal, figure out the steps, and deliver a result 2. agents don’t stop after one response. they keep running until the task is actually finishedno babysitting required 3. everything runs on a loop. they gather context, decide what to do, take an action, then repeat until done 4. the loop is the system. they look at files, tools, and the internet. decide the next step. execute and then feed that back into the next step. over and over until completion 5. the model is just one piece. gpt, claude, gemini are the reasoning layer. the key is model + loop + tools + context 6. mcp is how agents use tools. it connects things like browser, code, apis, and your internal software. once connected, the agent decides when to use them to get the job done 7. context beats prompt all day. you don't need to write perfect prompts. load your agent with context about your business, style, and goals and then simple instructions work 8. claude.md or agents.md is the onboarding doc it tells the agent who it is, how to behave, what it knows, and what tools it can use. this gets loaded every time before it starts 9. memory.md is how it improves. agents don’t remember by default. this file stores preferences, corrections, and patterns you tell the agent to update it, and it gets better over time 10. skills + harnesses make it usable. skills are reusable tasks like writing, research, analysis the harness is the environment like claude code or openclaw that runs everything. basiclaly, different interfaces, same system underneath this episode with remy on The Startup Ideas Podcast (SIP) 🧃 was one of the clearest ways of understanding a lot of the core concepts of ai agents could be the best beginners course for ai agents 58 mins. all free. no advertisers. i just want to see you build cool stuff. im rooting for you. send to a friend watch
58:55

AI AGENTS 101 (58 minute free masterclass) send this to anyone who wants to understand ai agents, claude skills, md files, how to get the most out of AI etc in plain english: 1. chat vs agents - chat models answer questions in a back and forth while agents take a goal, figure out the steps, and deliver a result 2. agents don’t stop after one response. they keep running until the task is actually finishedno babysitting required 3. everything runs on a loop. they gather context, decide what to do, take an action, then repeat until done 4. the loop is the system. they look at files, tools, and the internet. decide the next step. execute and then feed that back into the next step. over and over until completion 5. the model is just one piece. gpt, claude, gemini are the reasoning layer. the key is model + loop + tools + context 6. mcp is how agents use tools. it connects things like browser, code, apis, and your internal software. once connected, the agent decides when to use them to get the job done 7. context beats prompt all day. you don't need to write perfect prompts. load your agent with context about your business, style, and goals and then simple instructions work 8. claude.md or agents.md is the onboarding doc it tells the agent who it is, how to behave, what it knows, and what tools it can use. this gets loaded every time before it starts 9. memory.md is how it improves. agents don’t remember by default. this file stores preferences, corrections, and patterns you tell the agent to update it, and it gets better over time 10. skills + harnesses make it usable. skills are reusable tasks like writing, research, analysis the harness is the environment like claude code or openclaw that runs everything. basiclaly, different interfaces, same system underneath this episode with remy on The Startup Ideas Podcast (SIP) 🧃 was one of the clearest ways of understanding a lot of the core concepts of ai agents could be the best beginners course for ai agents 58 mins. all free. no advertisers. i just want to see you build cool stuff. im rooting for you. send to a friend watch

GREG ISENBERG

377,833 views • 6 months ago

BREAKING NEWS: Anthropic just dropped Claude Ops 4.5!! It is by FAR the best coding model I've ever used. We've been testing it internally Every 📧 for the last few days, and it is an absolute paradigm shift for any kind of coding task. It extends the horizon of what you can vibe code The current generation of new models—Anthropic’s Sonnet 4.5, Google’s Gemini 3, or OpenAI’s Codex Max 5.1—can all competently build a minimum viable product in one shot, or fix a highly technical bug autonomously. But eventually, if you kept pushing them to vibe code more, they’d start to trip over their own feet: The code would be convoluted and contradictory, and you’d get stuck in endless bugs. We have not found that limit yet with Opus 4.5—it seems to be able to vibe code forever. Takes working in parallel to a whole new level because it's far better at planning and coding, it can work with more autonomy—meaning you can do more in parallel without breaking anything . Kieran Klaassen worked on 11 different projects in six hours—and had good results on all of them. Great at design iteration Opus 4.5 is incredibly skilled at iterating through a design autonomously using an MCP like Playwright. previous models would lose the thread after a few cycles, or say a design was done when it wasn't. Opus 4.5 is incredible at autonomously iterating until a design is pixel perfect. we have a full 4,000 word vibe check on Every 📧 right now with everything we tested:
1:31

BREAKING NEWS: Anthropic just dropped Claude Ops 4.5!! It is by FAR the best coding model I've ever used. We've been testing it internally Every 📧 for the last few days, and it is an absolute paradigm shift for any kind of coding task. It extends the horizon of what you can vibe code The current generation of new models—Anthropic’s Sonnet 4.5, Google’s Gemini 3, or OpenAI’s Codex Max 5.1—can all competently build a minimum viable product in one shot, or fix a highly technical bug autonomously. But eventually, if you kept pushing them to vibe code more, they’d start to trip over their own feet: The code would be convoluted and contradictory, and you’d get stuck in endless bugs. We have not found that limit yet with Opus 4.5—it seems to be able to vibe code forever. Takes working in parallel to a whole new level because it's far better at planning and coding, it can work with more autonomy—meaning you can do more in parallel without breaking anything . Kieran Klaassen worked on 11 different projects in six hours—and had good results on all of them. Great at design iteration Opus 4.5 is incredibly skilled at iterating through a design autonomously using an MCP like Playwright. previous models would lose the thread after a few cycles, or say a design was done when it wasn't. Opus 4.5 is incredible at autonomously iterating until a design is pixel perfect. we have a full 4,000 word vibe check on Every 📧 right now with everything we tested:

Dan Shipper 📧

272,699 views • 10 months ago

Cursor vs Claude Code (day 3 of 30) today I re-tested Conductor and I don't want to overhype this but... what the actual fook 😮 this thing is amazing! closest things to the "perfect workflow" I have seen in a long time, and it's something I could see myself use full-time it's so minimal and the polar opposite of Cursor but it has all the essentials → Claude Code + Codex subscriptions (= unlimited usage) → great integration with Claude Code's plan mode → multiple tabs with multiple agents at the same time → can use multiple models, not just one → great file tree and diff viewer → best git worktree support I've ever seen there's honestly so much more, like a dedicated "review" button (which you can pair to your favorite review model, e.g. gpt-5.2-extra-high), buttons to create and merge pull requests, to launch your dev server, ... this solves literally EVERY fault I can think of when using Claude Code and/or Codex, combining both into one easy UX that not only does everything the tools can do, but actually adds useful stuff on top just 6 months ago I couldn't have dreamt about any this and the wildest part is that IT IS FREE (I'd literally pay for this UX lol) first time I'm genuinely enjoying myself since starting this challenge not saying it's a full Cursor replacement yet, it obviously isn't for a lot of people (no browser, no debug mode, no code indexing, etc) ... but this is the closest thing I've seen yet 👀 follow for day 4!
5:56

Cursor vs Claude Code (day 3 of 30) today I re-tested Conductor and I don't want to overhype this but... what the actual fook 😮 this thing is amazing! closest things to the "perfect workflow" I have seen in a long time, and it's something I could see myself use full-time it's so minimal and the polar opposite of Cursor but it has all the essentials → Claude Code + Codex subscriptions (= unlimited usage) → great integration with Claude Code's plan mode → multiple tabs with multiple agents at the same time → can use multiple models, not just one → great file tree and diff viewer → best git worktree support I've ever seen there's honestly so much more, like a dedicated "review" button (which you can pair to your favorite review model, e.g. gpt-5.2-extra-high), buttons to create and merge pull requests, to launch your dev server, ... this solves literally EVERY fault I can think of when using Claude Code and/or Codex, combining both into one easy UX that not only does everything the tools can do, but actually adds useful stuff on top just 6 months ago I couldn't have dreamt about any this and the wildest part is that IT IS FREE (I'd literally pay for this UX lol) first time I'm genuinely enjoying myself since starting this challenge not saying it's a full Cursor replacement yet, it obviously isn't for a lot of people (no browser, no debug mode, no code indexing, etc) ... but this is the closest thing I've seen yet 👀 follow for day 4!

Robin Ebers

40,764 views • 9 months ago

Anthropic's Claude Ai Agents Team just Educated how to build production AI agents in under 30 mins. For Free. From the engineers who built the stack. CANCEL Your Weekend Plans, and Learn to Build AI Agents Today. Bookmark it. Watch it. Build your first production agent this weekend. $5,000/month. $7,000/month. $12,000/month. People are building agents for clients and charging $$$ as Beginners. You're still stuck in the thinking about AI phase. This video fixes that tonight. Follow Himanshu Kumar for more high-signal content that actually moves your AI engineering career forward. ↓ Ivan Nardini runs Developer Relations for AI at Google Cloud. He just gave away the entire production agent stack in 30 minutes. This is the talk that separates people deploying AI agents that actually scale from people whose agents break the moment they leave localhost. Here's everything inside. I break down a production AI video like this every week. Follow Himanshu Kumar. ↓ The 4-part agent stack that actually scales. Most devs are duct-taping frameworks together and calling it an "AI agent." Ivan lays out the real stack: Agent Development Kit (ADK): open-source, code-first framework for building, evaluating, and deploying agents. Supports Claude models through Vertex AI directly. Model Context Protocol (MCP): lets your agent talk to any tool or data source with one standard. Vertex AI Agent Engine: managed platform for deploying, monitoring, and scaling agents in production. No DevOps headaches. Agent-to-Agent Protocol: open protocol so agents built on different frameworks can actually work together. This is the stack replacing every hacky agent setup in production right now. Full MCP + Claude breakdowns drop weekly on Himanshu Kumar. ↓ Building your first real agent. Ivan builds a birthday planner agent live. LLM Agent class. Name it. Define instructions. Pick the model. He uses Claude 3.7 Sonnet. You could use Opus 4.7 for better reasoning. Full agent built in minutes. Not weeks. Watch the build once and you'll never structure an agent the wrong way again. I post agent architectures people pay $500 courses to learn. Himanshu Kumar. ↓ Multi-agent systems without the chaos. Single agents are easy. Multi-agent systems are where 99% of builders fail. Ivan extends the birthday planner by: Adding a calendar service through MCP tools Creating an orchestrator agent to route requests between agents Handling state and context across agent handoffs This is production multi-agent architecture. Clean. Scalable. Debuggable. Most tutorials hand-wave this part. This one shows you every step. Multi-agent orchestration content drops weekly on Himanshu Kumar. ↓ Deployment without the DevOps nightmare. This is where most AI projects die. You build a cool agent locally. It works. You try to deploy it. Everything breaks. Vertex AI Agent Engine fixes this: Minimal code deployment Automatic monitoring of latency, CPU, and memory Built-in observability and logging No infrastructure setup needed You provide config and requirements. The platform handles the rest. This is how agents actually get to production. Deployment guides for Claude agents post every week. Himanshu Kumar. ↓ Agent-to-Agent Protocol: the future nobody's talking about. Most people don't know this exists yet. The A2A Protocol lets agents built in different frameworks communicate seamlessly. Your Claude agent. My LangChain agent. Someone else's CrewAI agent. All talking to each other. All solving parts of the same problem. All without custom integration code. This is the infrastructure layer of the coming AI economy. Getting in early on A2A Protocol is like getting in early on HTTP in 1995. A2A deep dive coming soon. Himanshu Kumar. ↓ 30 minutes from the team shipping this in production. You'll learn more from this than from 6 months of YouTube tutorials made by people who've never deployed an agent past localhost. People who watch this understand production AI agents at the architect level. People who skip it keep hacking together frameworks that break every time an API updates. Save the video. Watch it tonight. Build a real agent this weekend. Follow Himanshu Kumar for more high-signal content that actually moves your AI engineering career forward.
30:12

Anthropic's Claude Ai Agents Team just Educated how to build production AI agents in under 30 mins. For Free. From the engineers who built the stack. CANCEL Your Weekend Plans, and Learn to Build AI Agents Today. Bookmark it. Watch it. Build your first production agent this weekend. $5,000/month. $7,000/month. $12,000/month. People are building agents for clients and charging $$$ as Beginners. You're still stuck in the thinking about AI phase. This video fixes that tonight. Follow Himanshu Kumar for more high-signal content that actually moves your AI engineering career forward. ↓ Ivan Nardini runs Developer Relations for AI at Google Cloud. He just gave away the entire production agent stack in 30 minutes. This is the talk that separates people deploying AI agents that actually scale from people whose agents break the moment they leave localhost. Here's everything inside. I break down a production AI video like this every week. Follow Himanshu Kumar. ↓ The 4-part agent stack that actually scales. Most devs are duct-taping frameworks together and calling it an "AI agent." Ivan lays out the real stack: Agent Development Kit (ADK): open-source, code-first framework for building, evaluating, and deploying agents. Supports Claude models through Vertex AI directly. Model Context Protocol (MCP): lets your agent talk to any tool or data source with one standard. Vertex AI Agent Engine: managed platform for deploying, monitoring, and scaling agents in production. No DevOps headaches. Agent-to-Agent Protocol: open protocol so agents built on different frameworks can actually work together. This is the stack replacing every hacky agent setup in production right now. Full MCP + Claude breakdowns drop weekly on Himanshu Kumar. ↓ Building your first real agent. Ivan builds a birthday planner agent live. LLM Agent class. Name it. Define instructions. Pick the model. He uses Claude 3.7 Sonnet. You could use Opus 4.7 for better reasoning. Full agent built in minutes. Not weeks. Watch the build once and you'll never structure an agent the wrong way again. I post agent architectures people pay $500 courses to learn. Himanshu Kumar. ↓ Multi-agent systems without the chaos. Single agents are easy. Multi-agent systems are where 99% of builders fail. Ivan extends the birthday planner by: Adding a calendar service through MCP tools Creating an orchestrator agent to route requests between agents Handling state and context across agent handoffs This is production multi-agent architecture. Clean. Scalable. Debuggable. Most tutorials hand-wave this part. This one shows you every step. Multi-agent orchestration content drops weekly on Himanshu Kumar. ↓ Deployment without the DevOps nightmare. This is where most AI projects die. You build a cool agent locally. It works. You try to deploy it. Everything breaks. Vertex AI Agent Engine fixes this: Minimal code deployment Automatic monitoring of latency, CPU, and memory Built-in observability and logging No infrastructure setup needed You provide config and requirements. The platform handles the rest. This is how agents actually get to production. Deployment guides for Claude agents post every week. Himanshu Kumar. ↓ Agent-to-Agent Protocol: the future nobody's talking about. Most people don't know this exists yet. The A2A Protocol lets agents built in different frameworks communicate seamlessly. Your Claude agent. My LangChain agent. Someone else's CrewAI agent. All talking to each other. All solving parts of the same problem. All without custom integration code. This is the infrastructure layer of the coming AI economy. Getting in early on A2A Protocol is like getting in early on HTTP in 1995. A2A deep dive coming soon. Himanshu Kumar. ↓ 30 minutes from the team shipping this in production. You'll learn more from this than from 6 months of YouTube tutorials made by people who've never deployed an agent past localhost. People who watch this understand production AI agents at the architect level. People who skip it keep hacking together frameworks that break every time an API updates. Save the video. Watch it tonight. Build a real agent this weekend. Follow Himanshu Kumar for more high-signal content that actually moves your AI engineering career forward.

Himanshu Kumar

229,302 views • 5 months ago

An Anthropic researcher sat down next to me at a hackathon last week. Claude Opus 4.7 was running 4 agents on my laptop. Live. No manual input. She looked at the terminal and said: "What is this?" I showed her. 4 agents. 678 trades. 81% win rate. $16,200 last 30 days. She worked on the evals team. She'd never seen Claude pointed at 88 million on-chain trades. The setup is 3 public repos. All free. -> 88 million Polymarket trades. Every wallet. Every entry. Every exit. Every resolution. -> the framework that bridges Claude Opus 4.7 directly to live markets. Order placement, position tracking, exit timing. -> real-time WebSocket order book. Depth on both sides. No polling, no lag. Four agents. One loop. Agent 1 identifies which wallets win consistently across 88 million trades. Agent 2 reverse-engineers their entry timing. Agent 3 monitors order book volume spikes. Agent 4 sizes positions using Kelly. No overbet. Drawdown capped at 1.4% over 678 trades. 85% of windows get killed. No trade. The bot only enters when 3 signals align: -> Elite wallet consensus pointing the same direction. -> Price divergence with Binance and Coinbase both agreeing. -> Order book imbalance confirming the bias. Single-source price data was 57% accurate. All three together: 81%. Exit before resolution. Always. Losers hold to 0 or $1. The agents copy their exits. The agents don't gamble on that. My stack: Claude Opus 4.7 at $19/mo, VPS Hetzner at $4.99/mo, Everything else free. Total stats: $23.99/month. 30 days: 678 trades, 81% win rate, net +$16,200, max drawdown -1.2%, avg hold 4h 12m. She asked if Anthropic could test this internally. "We run Claude on benchmarks and evals. Nobody pointed it at a live market dataset with 88 million rows." Claude Opus 4.7 didn't need a system prompt. It read the wallet index, understood the signal structure, and wrote the combiner logic in one pass. The people who built the model hadn't thought to point it at this data. I had. Copy the live trades: -> all 4 agents run 24/7. The window is open right now. Save this, follow me and comment OPUS. I will send the guide to you.
0:26

An Anthropic researcher sat down next to me at a hackathon last week. Claude Opus 4.7 was running 4 agents on my laptop. Live. No manual input. She looked at the terminal and said: "What is this?" I showed her. 4 agents. 678 trades. 81% win rate. $16,200 last 30 days. She worked on the evals team. She'd never seen Claude pointed at 88 million on-chain trades. The setup is 3 public repos. All free. -> 88 million Polymarket trades. Every wallet. Every entry. Every exit. Every resolution. -> the framework that bridges Claude Opus 4.7 directly to live markets. Order placement, position tracking, exit timing. -> real-time WebSocket order book. Depth on both sides. No polling, no lag. Four agents. One loop. Agent 1 identifies which wallets win consistently across 88 million trades. Agent 2 reverse-engineers their entry timing. Agent 3 monitors order book volume spikes. Agent 4 sizes positions using Kelly. No overbet. Drawdown capped at 1.4% over 678 trades. 85% of windows get killed. No trade. The bot only enters when 3 signals align: -> Elite wallet consensus pointing the same direction. -> Price divergence with Binance and Coinbase both agreeing. -> Order book imbalance confirming the bias. Single-source price data was 57% accurate. All three together: 81%. Exit before resolution. Always. Losers hold to 0 or $1. The agents copy their exits. The agents don't gamble on that. My stack: Claude Opus 4.7 at $19/mo, VPS Hetzner at $4.99/mo, Everything else free. Total stats: $23.99/month. 30 days: 678 trades, 81% win rate, net +$16,200, max drawdown -1.2%, avg hold 4h 12m. She asked if Anthropic could test this internally. "We run Claude on benchmarks and evals. Nobody pointed it at a live market dataset with 88 million rows." Claude Opus 4.7 didn't need a system prompt. It read the wallet index, understood the signal structure, and wrote the combiner logic in one pass. The people who built the model hadn't thought to point it at this data. I had. Copy the live trades: -> all 4 agents run 24/7. The window is open right now. Save this, follow me and comment OPUS. I will send the guide to you.

slash1s

46,379 views • 5 months ago

glm 5.3 vs qwen 3.8 vs gemini 3.7 vs deepseek v4 flash four models designed and built three structures each on a physics-backed site, with no dimensions anywhere in the brief the setup: our own agent loop on OpenRouter, a construction site as the tool set – footings, walls, arches, roofs, scaffold, a lamp. the site enforces physics and nothing else: unsupported brick falls, a roof needs walls under it, a worker reaches 3.2 m above whatever he stands on, an arch needs centring until the keystone is set, concrete cures before it carries. no budget ceiling – material cost is tallied and reported, never blocked. tasks: 1. house – a plot and a palette, no plan. shape, height and material are the model's call 2. lighthouse – a headland cut by a gully, with a rock stack standing 30 m offshore. the lamp must burn, it must be the highest thing built, and the keeper must be able to walk to it 3. bridge – a river with one islet and banks at different heights. cross it however you want models: Z.ai glm 5.3 flash, Qwen qwen 3.8 flash, Google DeepMind gemini 3.7 flash, DeepSeek v4 flash vision all twelve objects were finished and signed off by the models themselves. tallest lighthouse is qwen's at 38.4 m, planted on the offshore stack with a bridge run out to it – the only model that read the site that way. deepseek signed off its bridge on an empty riverbed: 0 bricks, 107 minutes, $1.16m of material tallied - total cost, three builds #1 glm 5.3 flash – $0.201 #2 gemini 3.7 flash – $0.871 #3 qwen 3.8 flash – $1.058 #4 deepseek v4 flash – $1.567 - wall clock, three builds #1 gemini 3.7 flash – 91m #2 glm 5.3 flash – 228m #3 deepseek v4 flash – 502m #4 qwen 3.8 flash – 912m - total tokens #1 gemini 3.7 flash – 3,567,052 #2 glm 5.3 flash – 4,732,748 #3 qwen 3.8 flash – 13,469,333 #4 deepseek v4 flash – 18,230,076 - defects logged by the site #1 deepseek v4 flash – 59 #2 gemini 3.7 flash – 132 #3 glm 5.3 flash – 221 #4 qwen 3.8 flash – 350 - material tallied across three builds #1 gemini 3.7 flash – $359,884 #2 glm 5.3 flash – $583,358 #3 deepseek v4 flash – $1,327,484 #4 qwen 3.8 flash – $2,188,625 observations: • glm is the cheap one and nothing here is close – $0.201 for three buildings, $0.042 per million tokens, 6x under gemini's rate • what glm spends it on is bulk, not care: 166,228 bricks in one house and 156 defect weight, the worst single object in the set • gemini is the efficiency line – 91 minutes and 3.57m tokens for all three and an eighth of qwen's clock • gemini also builds the smallest of everything. its lighthouse is 22.5 m against qwen's 38.4, its house 6.9 m against 19.3 • qwen is the maximalist: 1.18m bricks, $2.19m of material, tallest on all three tasks, and 912 minutes – 15 hours – to get there conclusion: twelve finished objects for $3.80 all in, and a 7.8x price spread between the cheapest model and the priciest! follow thehype. for 24/7 ai news, analysis and breakdowns
1:09

glm 5.3 vs qwen 3.8 vs gemini 3.7 vs deepseek v4 flash four models designed and built three structures each on a physics-backed site, with no dimensions anywhere in the brief the setup: our own agent loop on OpenRouter, a construction site as the tool set – footings, walls, arches, roofs, scaffold, a lamp. the site enforces physics and nothing else: unsupported brick falls, a roof needs walls under it, a worker reaches 3.2 m above whatever he stands on, an arch needs centring until the keystone is set, concrete cures before it carries. no budget ceiling – material cost is tallied and reported, never blocked. tasks: 1. house – a plot and a palette, no plan. shape, height and material are the model's call 2. lighthouse – a headland cut by a gully, with a rock stack standing 30 m offshore. the lamp must burn, it must be the highest thing built, and the keeper must be able to walk to it 3. bridge – a river with one islet and banks at different heights. cross it however you want models: Z.ai glm 5.3 flash, Qwen qwen 3.8 flash, Google DeepMind gemini 3.7 flash, DeepSeek v4 flash vision all twelve objects were finished and signed off by the models themselves. tallest lighthouse is qwen's at 38.4 m, planted on the offshore stack with a bridge run out to it – the only model that read the site that way. deepseek signed off its bridge on an empty riverbed: 0 bricks, 107 minutes, $1.16m of material tallied - total cost, three builds #1 glm 5.3 flash – $0.201 #2 gemini 3.7 flash – $0.871 #3 qwen 3.8 flash – $1.058 #4 deepseek v4 flash – $1.567 - wall clock, three builds #1 gemini 3.7 flash – 91m #2 glm 5.3 flash – 228m #3 deepseek v4 flash – 502m #4 qwen 3.8 flash – 912m - total tokens #1 gemini 3.7 flash – 3,567,052 #2 glm 5.3 flash – 4,732,748 #3 qwen 3.8 flash – 13,469,333 #4 deepseek v4 flash – 18,230,076 - defects logged by the site #1 deepseek v4 flash – 59 #2 gemini 3.7 flash – 132 #3 glm 5.3 flash – 221 #4 qwen 3.8 flash – 350 - material tallied across three builds #1 gemini 3.7 flash – $359,884 #2 glm 5.3 flash – $583,358 #3 deepseek v4 flash – $1,327,484 #4 qwen 3.8 flash – $2,188,625 observations: • glm is the cheap one and nothing here is close – $0.201 for three buildings, $0.042 per million tokens, 6x under gemini's rate • what glm spends it on is bulk, not care: 166,228 bricks in one house and 156 defect weight, the worst single object in the set • gemini is the efficiency line – 91 minutes and 3.57m tokens for all three and an eighth of qwen's clock • gemini also builds the smallest of everything. its lighthouse is 22.5 m against qwen's 38.4, its house 6.9 m against 19.3 • qwen is the maximalist: 1.18m bricks, $2.19m of material, tallest on all three tasks, and 912 minutes – 15 hours – to get there conclusion: twelve finished objects for $3.80 all in, and a 7.8x price spread between the cheapest model and the priciest! follow thehype. for 24/7 ai news, analysis and breakdowns

thehype.

26,360 views • 1 month ago

ClawTeam v0.2.0 is here. One CLI to coordinate any coding agent — Claude Code, Codex, OpenClaw, nanobot, and more — into a self‑organizing swarm that plans, builds, and ships together. What's new in v0.2.0: 1) - Gource Visualization — Watch your agent swarm’s Git activity in real time. Clear. Visual. Instant. Run: clawteam board gource --live See every commit, branch, and merge as it happens. Track what each agent is doing. 2) - Runtime Profiles — A provider‑aware configuration system. Switch between Claude, Kimi, and Gemini anytime. No need to edit environment variables. Run clawteam profile wizard. Follow the interactive setup. Done in minutes. 3) - Git-Based Context — Full worktree isolation with built‑in conflict detection and change tracking. Each agent works on its own branch, and the leader can see everything clearly in one place. 4) - Stability & Hardening — Spawn/workspace conflict fixes, improved tmux integration, message normalization, P2P liveness with lease-based detection. This release is about making the foundation rock-solid. --------------------------------------------------------- To show what a coordinated agent swarm can actually do, we ran 1 Claude Code orchestrating 8 Claude Code agents to build a robotics simulation system optimized for Apple Silicon — from scratch. 8 hours. 300+ PRs. One running simulator. Check the result: --------------------------------------------------------- Huge thanks to the open-source community for the feedback, issues, and PRs that shaped this release. ClawTeam is built in the open because we believe multi-agent coordination should be a shared primitive, not a proprietary moat. Try it: pip install clawteam Docs: GitHub: #ClawTeam #nanobot #AIAgents #openclaw #ClaudeCode #Cursor
1:23

ClawTeam v0.2.0 is here. One CLI to coordinate any coding agent — Claude Code, Codex, OpenClaw, nanobot, and more — into a self‑organizing swarm that plans, builds, and ships together. What's new in v0.2.0: 1) - Gource Visualization — Watch your agent swarm’s Git activity in real time. Clear. Visual. Instant. Run: clawteam board gource --live See every commit, branch, and merge as it happens. Track what each agent is doing. 2) - Runtime Profiles — A provider‑aware configuration system. Switch between Claude, Kimi, and Gemini anytime. No need to edit environment variables. Run clawteam profile wizard. Follow the interactive setup. Done in minutes. 3) - Git-Based Context — Full worktree isolation with built‑in conflict detection and change tracking. Each agent works on its own branch, and the leader can see everything clearly in one place. 4) - Stability & Hardening — Spawn/workspace conflict fixes, improved tmux integration, message normalization, P2P liveness with lease-based detection. This release is about making the foundation rock-solid. --------------------------------------------------------- To show what a coordinated agent swarm can actually do, we ran 1 Claude Code orchestrating 8 Claude Code agents to build a robotics simulation system optimized for Apple Silicon — from scratch. 8 hours. 300+ PRs. One running simulator. Check the result: --------------------------------------------------------- Huge thanks to the open-source community for the feedback, issues, and PRs that shaped this release. ClawTeam is built in the open because we believe multi-agent coordination should be a shared primitive, not a proprietary moat. Try it: pip install clawteam Docs: GitHub: #ClawTeam #nanobot #AIAgents #openclaw #ClaudeCode #Cursor

Chao Huang

25,232 views • 6 months ago

elon musk grabbed the source code openai open-sourced by accident, rewrote it in rust over a weekend, and shipped it as a free coding agent that does everything $200/mo chatgpt pro does. why pay $200 to openai and $200 to claude when this runs for $8 the swarm above is one weekend of exactly that: thousands of agents pouring through four endpoints, three paid seats billing $1.80 a task while the free fork bills $0. musk co-founded openai, walked out, and when they left codex on github under a permissive license, he forked it, stamped grok on it, and gave it away what the free version does that the $200 seat charges for: the agent · openai's own engine -> it reads your repo, writes patches, runs your tests, and loops until they pass, exactly like codex -> because under the hood it is codex, just faster and free. you are paying $200 for the paid skin of a tool now sitting on github the license · apache-2.0, un-revocable -> free to use, free to fork, free to ship inside your own product with zero strings -> openai cannot pull it back. musk made sure the license is the kind that never expires the switch · one line, no new tools -> point it at any openai-compatible or claude-compatible endpoint, including an $8 kimi backend -> same terminal, same workflow, gpt-5.6 and opus 5 just quietly lose the seat the bill · $400 down to $8 -> chatgpt pro plus claude max is $400 a month. the free agent plus an $8 kimi key does the same daily work -> that is a 98% cut, built out of openai's own source code, handed to you by the guy suing them here is the part they will fight me on: openai did not lose this to a better model, they lost it to their own license and an enemy with a weekend free. the $200 was never the tool, it was the toll, and musk just put openai's own logo on the road around it drop your $400/mo ai stack to $8. the run above is openai's own agent, rewritten free, doing the job it bills $200 a month for. the full breakdown is in the article below
0:26

elon musk grabbed the source code openai open-sourced by accident, rewrote it in rust over a weekend, and shipped it as a free coding agent that does everything $200/mo chatgpt pro does. why pay $200 to openai and $200 to claude when this runs for $8 the swarm above is one weekend of exactly that: thousands of agents pouring through four endpoints, three paid seats billing $1.80 a task while the free fork bills $0. musk co-founded openai, walked out, and when they left codex on github under a permissive license, he forked it, stamped grok on it, and gave it away what the free version does that the $200 seat charges for: the agent · openai's own engine -> it reads your repo, writes patches, runs your tests, and loops until they pass, exactly like codex -> because under the hood it is codex, just faster and free. you are paying $200 for the paid skin of a tool now sitting on github the license · apache-2.0, un-revocable -> free to use, free to fork, free to ship inside your own product with zero strings -> openai cannot pull it back. musk made sure the license is the kind that never expires the switch · one line, no new tools -> point it at any openai-compatible or claude-compatible endpoint, including an $8 kimi backend -> same terminal, same workflow, gpt-5.6 and opus 5 just quietly lose the seat the bill · $400 down to $8 -> chatgpt pro plus claude max is $400 a month. the free agent plus an $8 kimi key does the same daily work -> that is a 98% cut, built out of openai's own source code, handed to you by the guy suing them here is the part they will fight me on: openai did not lose this to a better model, they lost it to their own license and an enemy with a weekend free. the $200 was never the tool, it was the toll, and musk just put openai's own logo on the road around it drop your $400/mo ai stack to $8. the run above is openai's own agent, rewritten free, doing the job it bills $200 a month for. the full breakdown is in the article below

starmex

111,684 views • 1 month ago

Cerebras inference is very fast. So fast that it changes how we think about configuring our LLMs for voice agent use cases. Kimi K2.6 is a 1T parameter reasoning model that Cerebras serves at 650 - 1,000 tokens per second (end-to-end throughput), with time to first token metrics as low as 150ms (latency). These numbers are two to three times faster than other similarly capable models. The biggest lever we get from this kind of speed is that we can use the model in reasoning mode, and still have excellent "time to first non-thinking token." This solves a big pain point we have in 2026 for voice agent use cases. Almost all recent innovation in post-training has focused on making models good at reasoning ("test time compute"). This is great, but it makes the user-facing model latency much, much slower. Which is a problem for conversational voice agents. We can run Kimi K2.6 with reasoning turned on, and get responses faster than other models produce with reasoning disabled. On my 30-turn voice agent benchmark, Kimi K2.6 with reasoning enabled ties GPT 5.1 and Haiku 4.5 with reasoning disabled, and is still about 200ms seconds faster! On my primary task agent benchmark, Kimi K2.6 is now the #2 model. It ranks just behind Gemini 3.5 Flash in "high" reasoning mode, and tied with GLM 5, Sonnet 4.6, and GPT 5.4 with reasoning set to "low." But Kimi K2.6 completes each turn in the agent loop in under 500ms. The other four models are all at least 3x slower. (Models only qualify for this benchmark if they can complete task turns at a P50 <4s.) A couple of other things that this speed buys us, for production voice agents: - Tool calls happen fast enough that we don't have to work around tool call latency in our pipeline design. - We can prompt the model to output structured data at the beginning of a response, followed by plain text for voice generation. This opens up possibilities like asking the model to do complex classification/generation tasks that influence the rest of the pipeline. For example, the model could create a detailed style prompt for a steerable TTS model, for each individual conversation turn. And, of course, you can use Kimi K2.6 with reasoning turned off. Cerebras calls this "instant" mode. Here's a video of a Cerebras Kimi K2.6 voice agent with voice-to-voice response time, measured at the client, under 500ms. This is the true response latency as perceived by the user, including all network and audio codec overhead, transcription and turn detection, Kimi K2.6 token generation, and voice generation. 500ms is, effectively, instant. So the Cerebras naming for this mode is a propos. :-)
0:40

Cerebras inference is very fast. So fast that it changes how we think about configuring our LLMs for voice agent use cases. Kimi K2.6 is a 1T parameter reasoning model that Cerebras serves at 650 - 1,000 tokens per second (end-to-end throughput), with time to first token metrics as low as 150ms (latency). These numbers are two to three times faster than other similarly capable models. The biggest lever we get from this kind of speed is that we can use the model in reasoning mode, and still have excellent "time to first non-thinking token." This solves a big pain point we have in 2026 for voice agent use cases. Almost all recent innovation in post-training has focused on making models good at reasoning ("test time compute"). This is great, but it makes the user-facing model latency much, much slower. Which is a problem for conversational voice agents. We can run Kimi K2.6 with reasoning turned on, and get responses faster than other models produce with reasoning disabled. On my 30-turn voice agent benchmark, Kimi K2.6 with reasoning enabled ties GPT 5.1 and Haiku 4.5 with reasoning disabled, and is still about 200ms seconds faster! On my primary task agent benchmark, Kimi K2.6 is now the #2 model. It ranks just behind Gemini 3.5 Flash in "high" reasoning mode, and tied with GLM 5, Sonnet 4.6, and GPT 5.4 with reasoning set to "low." But Kimi K2.6 completes each turn in the agent loop in under 500ms. The other four models are all at least 3x slower. (Models only qualify for this benchmark if they can complete task turns at a P50 <4s.) A couple of other things that this speed buys us, for production voice agents: - Tool calls happen fast enough that we don't have to work around tool call latency in our pipeline design. - We can prompt the model to output structured data at the beginning of a response, followed by plain text for voice generation. This opens up possibilities like asking the model to do complex classification/generation tasks that influence the rest of the pipeline. For example, the model could create a detailed style prompt for a steerable TTS model, for each individual conversation turn. And, of course, you can use Kimi K2.6 with reasoning turned off. Cerebras calls this "instant" mode. Here's a video of a Cerebras Kimi K2.6 voice agent with voice-to-voice response time, measured at the client, under 500ms. This is the true response latency as perceived by the user, including all network and audio codec overhead, transcription and turn detection, Kimi K2.6 token generation, and voice generation. 500ms is, effectively, instant. So the Cerebras naming for this mode is a propos. :-)

kwindla

40,593 views • 4 months ago