Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

Introducing Launch Log 1 – a quarter’s worth of product updates: ✨Agent Prompts 📋Rules 🧠Codebase as context 📦Memory improvements ⭐New “update Warp” UI 🪟Warp on Windows!

Warp

57,366 subscribers

642,049 Aufrufe • vor 1 Jahr •via X (Twitter)

Wissenschaft & Technologie

Anya Rossi• Live Now

Private livecam show

20 Kommentare

Profilbild von Warp

Warpvor 1 Jahr

Prompts let you save the tasks you keep asking agents to do and share them with peers. We’ve curated some of our favorites on this site 👉 And yes, we did build the site with Agents in Warp.

Profilbild von Warp

Warpvor 1 Jahr

Rules let you save bits of context to be used in all Agent requests. What tools you use, style norms, and details about your dev environment.

Profilbild von Warp

Warpvor 1 Jahr

Warp will now scan your repo to give better responses, tailored to your current codebase. Note: We do not store any code on our servers.

Profilbild von Warp

Warpvor 1 Jahr

There’s a lot more to talk about — check out our blog for the longer list!

Profilbild von Catalin

Catalinvor 1 Jahr

Warp is GOATED! I’m converting all my common use cases to Agent Prompts. 🔥 I just shared 3 examples in this post:

Profilbild von Kunal Kushwaha

Kunal Kushwahavor 1 Jahr

Have to try this today after my meetup!

Profilbild von Warp

Warpvor 1 Jahr

Let us know what you think!

Profilbild von Erik Eliason

Erik Eliasonvor 1 Jahr

Agent Prompts are a superpower! Save, reuse, and share — makes agents way more powerful.

Profilbild von Warp

Warpvor 1 Jahr

to see more!

Profilbild von Csaba Kissi

Csaba Kissivor 1 Jahr

Agent prompts are game-changing

Profilbild von Sumanth

Sumanthvor 1 Jahr

Congrats on the launch team!🫡

Profilbild von Warp

Warpvor 1 Jahr

Thanks! Have a favorite feature?

Profilbild von techfren

techfrenvor 1 Jahr

Prompts are going to be replacing a bunch of micro SaaS apps

Profilbild von Warp

Warpvor 1 Jahr

We agree

Profilbild von Aadit Sheth

Aadit Shethvor 1 Jahr

looks so cool!

Profilbild von Warp

Warpvor 1 Jahr

Thank you!!

Profilbild von Melvin Vivas

Melvin Vivasvor 1 Jahr

wow, very useful feature!

Profilbild von Warp

Warpvor 1 Jahr

Thanks, Melvin!

Profilbild von Christian Lempa

Christian Lempavor 1 Jahr

Wow, these features are super cool and helpful!

Profilbild von CodeRabbit

CodeRabbitvor 1 Jahr

AI-first pull request reviewer with context-aware feedback, line-by-line code suggestions, and real-time chat.

Ähnliche Videos

We've added a Merchant 🛒 and a new Consumables System 🧪 What's new? 🖱️ Drag & drop your UI windows anywhere on your screen. ⚡ Quick consumable slots (1, 2, 3). ⚔️Buff System ✨ Several small UX improvements. This update opens an even wider window for future updates and exciting new features.

We've added a Merchant 🛒 and a new Consumables System 🧪 What's new? 🖱️ Drag & drop your UI windows anywhere on your screen. ⚡ Quick consumable slots (1, 2, 3). ⚔️Buff System ✨ Several small UX improvements. This update opens an even wider window for future updates and exciting new features.

The Islands

22,990 Aufrufe • vor 2 Monaten

Today is our biggest product launch yet. Announcing Warp Fabric. The first AI-native employee management platform with IT built in. Warp Fabric is Warp's IT product suite, built natively inside the same platform that runs your payroll, onboarding, and HRIS. No manual checklists. No third party dashboards. Setup that takes 10 minutes. Warp Fabric has three products: Google Workspace automation: accounts created on hire, suspended on termination, files transferred, sessions killed. App provisioning: every app your new hire needs ready before day one, access revoked the moment they leave, across 6,500+ apps. Warp Device Agent (MDM): continuous device monitoring, policies enforced on every connection, built in Rust, SOC 2 compliant out of the box, macOS and Windows on the same codebase from day one. You hire someone in Warp, every account, every app, every device is handled. You terminate someone, everything stops. Every competitor either bolts IT onto their HRIS through integrations, or resells a third-party tool under their own brand. We built ours differently. IT is not a separate system that receives signals from HR. It is one more thing the employee context graph handles automatically. AI-native architecture means the system understands employment events and acts on them. It does not wait for configured rules to fire in sequence. When someone is hired, provisioning is not triggered: it happens, as part of the same operation. Security incidents are rising. Auditors are showing up at companies asking for device compliance evidence most teams can't produce. In Warp, every policy check, every drift event is already logged. The answer is in Warp. Not a spreadsheet. Some of our fastest growing customers are already running Warp Fabric in their most critical workflows. This is a major step in our mission to build the most powerful and delightful employee management platform for ambitious companies. And we're just getting started. Warp Fabric is available to all Warp customers today.

Ayush S

3,783,825 Aufrufe • vor 4 Monaten

Tainted Grail: The Fall of Avalon, version 1.20, is now live on Steam! 🔥 This update is huge and meaningful. We're introducing: - several balance changes (some of them being optional), - new content & Act 3 improvements, - various fixes and quality-of-life improvements, - a ton of audio improvements, - console improvements (40 FPS support & fixed PlayStation’s saving failed issue). (Please note that, as per usual, consoles will need a few extra days to process and approve the update before it goes live.) Check out the full list of changes below. It's worth the read! #taintedgrail #rpg #update #patchnotes

Tainted Grail: The Fall of Avalon, version 1.20, is now live on Steam! 🔥 This update is huge and meaningful. We're introducing: - several balance changes (some of them being optional), - new content & Act 3 improvements, - various fixes and quality-of-life improvements, - a ton of audio improvements, - console improvements (40 FPS support & fixed PlayStation’s saving failed issue). (Please note that, as per usual, consoles will need a few extra days to process and approve the update before it goes live.) Check out the full list of changes below. It's worth the read! #taintedgrail #rpg #update #patchnotes

Tainted Grail

85,968 Aufrufe • vor 5 Monaten

1/ Context Lens, new in Tlon v9.2.0. Everything your agent has ever done, stored durably on your personal server. Every message, every tool call. Debugging an agent usually means ssh-ing into a server and grepping logs. Now it's a tap in Tlon Messenger's UI. Tap the badge under a bot's reply to open the run: the context it read, each tool call with timing, the model that answered, and a persistence log of everything it wrote.

1/ Context Lens, new in Tlon v9.2.0. Everything your agent has ever done, stored durably on your personal server. Every message, every tool call. Debugging an agent usually means ssh-ing into a server and grepping logs. Now it's a tap in Tlon Messenger's UI. Tap the badge under a bot's reply to open the run: the context it read, each tool call with timing, the model that answered, and a persistence log of everything it wrote.

Tlon

10,680 Aufrufe • vor 1 Monat

10 GitHub repos with ⭐ 450k+ combined stars that stop your agent from wasting tokens 1. caveman 98k⭐ - makes your AI agent talk like a caveman. Same answers, 65% fewer output tokens. 2. rtk 76k⭐ - CLI proxy that filters git/test/docker output before it hits your context. Up to 90% less bash noise. 3. mem0 63k⭐ - memory layer for AI agents. Stops your agent from re-explaining the same context every session. 4. context7 60.7k⭐ - pulls exact, up-to-date docs into context instead of your agent guessing and burning tokens on retries. 5. aider 48k⭐ - builds a map of your repo instead of loading the whole codebase into context. 6. code-review-graph 30k⭐ - graphs your codebase so your agent reads 2-3 files instead of the whole repo. Up to 376x fewer tokens per question. 7. serena 28k⭐ - gives agents symbol-level code access instead of reading full files line by line. 8. repomix 28k⭐ - compresses your entire repo into an AI-friendly format. ~70% token reduction. 9. ccusage 18k⭐ - shows you exactly where your Claude Code sessions are burning tokens. 10. GPTCache 8k⭐ - semantic cache for LLM calls. Stop paying twice for the same question. Save this before your next agent session burns through your subscription 👇

10 GitHub repos with ⭐ 450k+ combined stars that stop your agent from wasting tokens 1. caveman 98k⭐ - makes your AI agent talk like a caveman. Same answers, 65% fewer output tokens. 2. rtk 76k⭐ - CLI proxy that filters git/test/docker output before it hits your context. Up to 90% less bash noise. 3. mem0 63k⭐ - memory layer for AI agents. Stops your agent from re-explaining the same context every session. 4. context7 60.7k⭐ - pulls exact, up-to-date docs into context instead of your agent guessing and burning tokens on retries. 5. aider 48k⭐ - builds a map of your repo instead of loading the whole codebase into context. 6. code-review-graph 30k⭐ - graphs your codebase so your agent reads 2-3 files instead of the whole repo. Up to 376x fewer tokens per question. 7. serena 28k⭐ - gives agents symbol-level code access instead of reading full files line by line. 8. repomix 28k⭐ - compresses your entire repo into an AI-friendly format. ~70% token reduction. 9. ccusage 18k⭐ - shows you exactly where your Claude Code sessions are burning tokens. 10. GPTCache 8k⭐ - semantic cache for LLM calls. Stop paying twice for the same question. Save this before your next agent session burns through your subscription 👇

unicode

66,891 Aufrufe • vor 15 Tagen

I miss doing these video updates for Chatbase. This is our first update in the great lock in of q4. ✨New onboarding flow From 1 link to an AI agent, trained on your site, with your logo, brand colors, and a custom prompt in less than a minute.

Yasser

32,856 Aufrufe • vor 11 Monaten

New course: Agent Memory: Building Memory-Aware Agents, built in partnership with Oracle and taught by Richmond Alake and Nacho Martínez. Many agents work well within a single session but their memory resets once the session ends. Consider a research agent working on dozens of papers across multiple days: without memory, it has no way to store and retrieve what it learned across sessions. This short course teaches you to build a memory system that enables agents to persist memory and thereby learn across sessions. You'll design a Memory Manager that handles different memory types, implement semantic tool retrieval that scales without bloating the context, and build write-back pipelines that let your agent autonomously update and refine what it knows over time. Skills you'll gain: - Build persistent memory stores for different agent memory types - Implement a Memory Manager that orchestrates how your agent reads, writes, and retrieves memory - Treat tools as procedural memory and retrieve only relevant ones at inference time using semantic search Join and learn to build agents that remember and improve over time!

New course: Agent Memory: Building Memory-Aware Agents, built in partnership with Oracle and taught by Richmond Alake and Nacho Martínez. Many agents work well within a single session but their memory resets once the session ends. Consider a research agent working on dozens of papers across multiple days: without memory, it has no way to store and retrieve what it learned across sessions. This short course teaches you to build a memory system that enables agents to persist memory and thereby learn across sessions. You'll design a Memory Manager that handles different memory types, implement semantic tool retrieval that scales without bloating the context, and build write-back pipelines that let your agent autonomously update and refine what it knows over time. Skills you'll gain: - Build persistent memory stores for different agent memory types - Implement a Memory Manager that orchestrates how your agent reads, writes, and retrieves memory - Treat tools as procedural memory and retrieve only relevant ones at inference time using semantic search Join and learn to build agents that remember and improve over time!

Andrew Ng

161,236 Aufrufe • vor 5 Monaten

New short course: Long-Term Agentic Memory with LangGraph. Learn to build an agent with long-term memory in this course developed in collaboration with taught by its Co-Founder and CEO, Harrison Chase! Personal assistance and productivity tasks have become important use cases for agents. An important feature of an AI assistant, such as a coding or calendar assistant, is its ability to keep improving over time from its experience. Agent memory is the key capability that enables this. To add memory to an agent, you must first figure out what to store and what to retrieve when it is time to use the information. Additionally, you’ll have to decide when to update the stored information. For example, you might update in each iteration loop of the agent or perform updates in the background, with a helper agent. In this course, you will learn a mental framework to build agents with long-term memory. You'll create a useful email assistant that can respond, ignore, and notify using writing, scheduling, and memory-management tools. You’ll develop your agent's memory by adding facts to its memory store, provide examples to learn the user's preferences, and optimize system prompts to evolve instructions based on previous responses. In detail, you’ll: - Learn how the three types of memory--semantic, episodic, and procedural–and the two update mechanisms–via hot path and in the background–apply to your agents. - Build an email agent with writing, scheduling, and availability tools, along with a router that triages incoming email and handles it accordingly by ignoring, responding, or notifying the user. - Add tools to your email agent that allow it to operate on semantic memory by learning facts about the user, storing them in a long-term memory store, and searching over them in future interactions. - Incorporate episodic memory, in the form of few-shot examples, in the triage step of your agents to help them learn and update user preferences. - Add procedural memory as system prompts, optimized with feedback to improve the instructions the agent follows. Learn how to approach memory in agents, and start building agents with long-term memory with LangGraph! Please sign up here:

New short course: Long-Term Agentic Memory with LangGraph. Learn to build an agent with long-term memory in this course developed in collaboration with taught by its Co-Founder and CEO, Harrison Chase! Personal assistance and productivity tasks have become important use cases for agents. An important feature of an AI assistant, such as a coding or calendar assistant, is its ability to keep improving over time from its experience. Agent memory is the key capability that enables this. To add memory to an agent, you must first figure out what to store and what to retrieve when it is time to use the information. Additionally, you’ll have to decide when to update the stored information. For example, you might update in each iteration loop of the agent or perform updates in the background, with a helper agent. In this course, you will learn a mental framework to build agents with long-term memory. You'll create a useful email assistant that can respond, ignore, and notify using writing, scheduling, and memory-management tools. You’ll develop your agent's memory by adding facts to its memory store, provide examples to learn the user's preferences, and optimize system prompts to evolve instructions based on previous responses. In detail, you’ll: - Learn how the three types of memory--semantic, episodic, and procedural–and the two update mechanisms–via hot path and in the background–apply to your agents. - Build an email agent with writing, scheduling, and availability tools, along with a router that triages incoming email and handles it accordingly by ignoring, responding, or notifying the user. - Add tools to your email agent that allow it to operate on semantic memory by learning facts about the user, storing them in a long-term memory store, and searching over them in future interactions. - Incorporate episodic memory, in the form of few-shot examples, in the triage step of your agents to help them learn and update user preferences. - Add procedural memory as system prompts, optimized with feedback to improve the instructions the agent follows. Learn how to approach memory in agents, and start building agents with long-term memory with LangGraph! Please sign up here:

Andrew Ng

131,850 Aufrufe • vor 1 Jahr

WARP SPEED: EPISODE 3 - starring Dada, CEO of Sorce Jobs Dada You only get to live once. So you have to use the best software you can. Dada, daniel ajayi and david built Sorce (YC F25) into Tinder for Jobs—20 million swipes, companies like SpaceX, Anduril, Stripe, DoorDash, and Lyft hiring through their platform. 2.7 million video views. Got into YC three days after the deadline. Dada didn't plan to start a company. He built Sorce to apply to jobs faster for himself. Then SpaceX hired someone through Sorce. Then Anduril did too. "Our biggest fear was: are you just going to destroy the industry? Then Anduril hired someone from Sorce, and SpaceX hired someone from Sorce. I was like, okay, they actually like us." Now Sorce is contributing to the economy: - 20 million swipes in total - Every major tech company has interviewed or hired someone from Sorce - On track to contribute to 1% of all applications in the world - Built the first version in 2-3 weeks, launched the app in 7 weeks - Team of 5 moving at Gabriella Warp Speed In this conversation: (0:00) "I didn't want to start a company. I just wanted to apply to jobs faster" (0:36) Why Y Combinator: Working with world-class people like the founder of Google Photos David Lieb (0:48) The problem: Applying to jobs is painfully repetitive (01:07) Viral launch: 2.7M views—LinkedIn loved it, Twitter said "you're going to destroy the world" (01:45) From 30,000 swipes to 20M: On track for 1% of all applications in the world (02:45) Building the team: 5 people shipping at Warp speed (03:45) Shipping velocity: Built Tinder for Jobs in 7 weeks (04:20) Switching from Rippling: "The UI was very confusing. I paid a contractor twice" (04:33) Switching to Warp: “Warp is a breath of fresh air, it’s like moving from Teams to Slack. The design is just really beautiful” (04:58) Setting the bar: "You want to be inspired by everything you use. It also helps you as a founder to know where the bar is. I’m building a B2B product soon and it should look like Warp and it should be as simple as warp”

WARP SPEED: EPISODE 3 - starring Dada, CEO of Sorce Jobs Dada You only get to live once. So you have to use the best software you can. Dada, daniel ajayi and david built Sorce (YC F25) into Tinder for Jobs—20 million swipes, companies like SpaceX, Anduril, Stripe, DoorDash, and Lyft hiring through their platform. 2.7 million video views. Got into YC three days after the deadline. Dada didn't plan to start a company. He built Sorce to apply to jobs faster for himself. Then SpaceX hired someone through Sorce. Then Anduril did too. "Our biggest fear was: are you just going to destroy the industry? Then Anduril hired someone from Sorce, and SpaceX hired someone from Sorce. I was like, okay, they actually like us." Now Sorce is contributing to the economy: - 20 million swipes in total - Every major tech company has interviewed or hired someone from Sorce - On track to contribute to 1% of all applications in the world - Built the first version in 2-3 weeks, launched the app in 7 weeks - Team of 5 moving at Gabriella Warp Speed In this conversation: (0:00) "I didn't want to start a company. I just wanted to apply to jobs faster" (0:36) Why Y Combinator: Working with world-class people like the founder of Google Photos David Lieb (0:48) The problem: Applying to jobs is painfully repetitive (01:07) Viral launch: 2.7M views—LinkedIn loved it, Twitter said "you're going to destroy the world" (01:45) From 30,000 swipes to 20M: On track for 1% of all applications in the world (02:45) Building the team: 5 people shipping at Warp speed (03:45) Shipping velocity: Built Tinder for Jobs in 7 weeks (04:20) Switching from Rippling: "The UI was very confusing. I paid a contractor twice" (04:33) Switching to Warp: “Warp is a breath of fresh air, it’s like moving from Teams to Slack. The design is just really beautiful” (04:58) Setting the bar: "You want to be inspired by everything you use. It also helps you as a founder to know where the bar is. I’m building a B2B product soon and it should look like Warp and it should be as simple as warp”

Ayush S

20,553 Aufrufe • vor 9 Monaten

Announcing Warp 2.0 When we launched Warp last year, our goal was simple: take care of state tax compliance so founders never have to log into another government website again. Turns out, founders wanted that. Since then: - Revenue has grown over 20x - We now process tens of millions of dollars in payroll every month - And some of the best and fastest growing startups today like Bland, @JuliusAI_ , Reducto run their companies on Warp. We support companies with just 2 cofounders all the way to companies with 200 employees+ and along the way, it became clear that parts of our platform needed to evolve. So we’ve rebuilt the product from the ground up.Warp 2.0 has a brand new design that makes it easier to use, especially as your team grows. We cleaned up the layout, made it simpler to find what you need, and added smart tools to help you get things done faster. Whether you're paying 5 people or 500, everything is easier to see, manage, and approve. Let me show you.

Ayush S

28,401 Aufrufe • vor 1 Jahr

SpaceFi 2.0 Launching on ZKsync ! ⭐ As one of the earliest and most active projects in #zkSync ecosystem, more than 320k users have been active in our DEX+NFT+Farm+Spacebase. Love to all❤️ Now it's time to unveil the brand new 2.0 with improved product, tokenomics and UI, which means a new era. Try the new features such as sLP, Boosted Farm, Paymaster and new STAR / Planet NFT. With zkSync and Layer2 ecosystem thrives, we'll keep building and giving back more benefits to the community. Join us into SpaceFi 2.0. Details about 2.0 Migration and Updates: Key Steps 👇 1/5

SpaceFi (∎, ∆) 🦇🔊

52,353 Aufrufe • vor 2 Jahren

📢 'Leveraged Volatility Farming' (LVF) Public Beta Launch Announcement: 🚀 LVF contracts, New UI, $140k Contract Bug Bounty, and $20k Community UI Bounty Program 🚀 After months of planning and hard work, we’re finally sharing the official timeline for LVF and our new UI—coming together for a groundbreaking release! Here's what's on the horizon: 🗓 Launch Date: Week commencing January 20, 2025 🫛 LVF Beta Launch: - Open and manage LVF positions including lending, borrowing, and providing liquidity - Utilize our groundbreaking Self-Lending Pods and Metavaults, unlocking more flexible and capital-efficient strategies ✨ New UI: - Fast load times for seamless interactions - A more intuitive design to simplify protocol navigation - Meticulously redesigned user journeys to enhance the overall UX 💥 Security: A $140k private smart contract bug bounty program will run concurrently as we continue to place security as our #1 priority. In the coming weeks, we'll share more details about the launch process, the $140k Smart Contract Bug Bounty, and the $20k UI Bounty Program. Stay tuned—this is just the beginning! 🫛🚀 $PEAS

📢 'Leveraged Volatility Farming' (LVF) Public Beta Launch Announcement: 🚀 LVF contracts, New UI, $140k Contract Bug Bounty, and $20k Community UI Bounty Program 🚀 After months of planning and hard work, we’re finally sharing the official timeline for LVF and our new UI—coming together for a groundbreaking release! Here's what's on the horizon: 🗓 Launch Date: Week commencing January 20, 2025 🫛 LVF Beta Launch: - Open and manage LVF positions including lending, borrowing, and providing liquidity - Utilize our groundbreaking Self-Lending Pods and Metavaults, unlocking more flexible and capital-efficient strategies ✨ New UI: - Fast load times for seamless interactions - A more intuitive design to simplify protocol navigation - Meticulously redesigned user journeys to enhance the overall UX 💥 Security: A $140k private smart contract bug bounty program will run concurrently as we continue to place security as our #1 priority. In the coming weeks, we'll share more details about the launch process, the $140k Smart Contract Bug Bounty, and the $20k UI Bounty Program. Stay tuned—this is just the beginning! 🫛🚀 $PEAS

Peapods Finance

72,649 Aufrufe • vor 1 Jahr

A DEVELOPER CONNECTED CLAUDE CODE TO OBSIDIAN SO HIS AI AGENT WOULD STOP FORGETTING THE PROJECT EVERY MORNING. Every coding session used to start the same way. Claude would understand the repo, fix the bug, explain the architecture, and then the moment the session ended, all of that context disappeared. Same codebase. Same decisions. Same architecture. Same mistakes repeated again. So he added a memory layer. Instead of treating Claude Code like a smart terminal, he connected it to a local Obsidian vault through MCP. Now Claude can read the repo, open the vault, create notes, link concepts, and write important decisions back into the system. When it studies the codebase, it does not just answer once and forget. It creates notes for the major services, maps how the architecture works, links auth to the database, connects APIs to storage, and records why certain migrations or design choices exist. Obsidian becomes the project graph. Now when he asks why something was built a certain way, Claude does not guess from the current prompt. It reads the decision notes. When he starts a new branch, Claude checks the active context file. When the work is done, it updates what changed, what is blocked, and what the next agent needs to know before touching the repo. That is the real loop: read context, write code, capture decisions, update memory. Most people are still using AI coding tools like disposable chat windows. Ask, patch, close, forget. This setup turns Claude Code into infrastructure. The repo gets a memory layer that survives every session, and multiple AI agents can work from the same project map without stepping on each other. The unlock is not better prompting. The unlock is giving the agent somewhere to remember what it already learned.

A DEVELOPER CONNECTED CLAUDE CODE TO OBSIDIAN SO HIS AI AGENT WOULD STOP FORGETTING THE PROJECT EVERY MORNING. Every coding session used to start the same way. Claude would understand the repo, fix the bug, explain the architecture, and then the moment the session ended, all of that context disappeared. Same codebase. Same decisions. Same architecture. Same mistakes repeated again. So he added a memory layer. Instead of treating Claude Code like a smart terminal, he connected it to a local Obsidian vault through MCP. Now Claude can read the repo, open the vault, create notes, link concepts, and write important decisions back into the system. When it studies the codebase, it does not just answer once and forget. It creates notes for the major services, maps how the architecture works, links auth to the database, connects APIs to storage, and records why certain migrations or design choices exist. Obsidian becomes the project graph. Now when he asks why something was built a certain way, Claude does not guess from the current prompt. It reads the decision notes. When he starts a new branch, Claude checks the active context file. When the work is done, it updates what changed, what is blocked, and what the next agent needs to know before touching the repo. That is the real loop: read context, write code, capture decisions, update memory. Most people are still using AI coding tools like disposable chat windows. Ask, patch, close, forget. This setup turns Claude Code into infrastructure. The repo gets a memory layer that survives every session, and multiple AI agents can work from the same project map without stepping on each other. The unlock is not better prompting. The unlock is giving the agent somewhere to remember what it already learned.

DegenCalls

20,124 Aufrufe • vor 1 Monat

Introducing Composer — the first AI Agent for document processing. Get to production-grade accuracy, autonomously in minutes. In our early beta, some teams hit 99% accuracy on complex document tasks in under 10 minutes. Composer is an agent built to optimize schemas the same way a human would (but way faster). Instead of tuning prompts by hand, you point Composer at your eval set inside Extend. Composer will: - analyze where your schema falls short - propose targeted improvements - run multiple experiments in parallel - surface diffs, accuracy gains, and traces behind each change With this launch, Extend is the only product on the market that helps you reach production-grade accuracy this fast. Composer is live for all Extend customers today! Try it out at the link in comments below.

Introducing Composer — the first AI Agent for document processing. Get to production-grade accuracy, autonomously in minutes. In our early beta, some teams hit 99% accuracy on complex document tasks in under 10 minutes. Composer is an agent built to optimize schemas the same way a human would (but way faster). Instead of tuning prompts by hand, you point Composer at your eval set inside Extend. Composer will: - analyze where your schema falls short - propose targeted improvements - run multiple experiments in parallel - surface diffs, accuracy gains, and traces behind each change With this launch, Extend is the only product on the market that helps you reach production-grade accuracy this fast. Composer is live for all Extend customers today! Try it out at the link in comments below.

Kushal Byatnal

75,152 Aufrufe • vor 11 Monaten

Mastering OpenAI's NEW AI AGENT builder in 26 mins: – start with one simple flow (lead capture or faq bot) before adding logic complexity. – use vector stores sparingly - too much context slows performance. – name your nodes clearly so you can debug fast. – integrate with chatkit early if you plan to embed it on a site. – test edge cases (misclassifications, loops, context drops). – think in outcomes, not features - what task can an agent take off your plate today? – treat this as a sandbox, not production - learn what’s possible before scaling. ideas on what you can build (to get your creative juices flowing): – lead qualification → build a multi-agent flow that classifies visitors as hot, warm, or cold, then sends data straight into hubspot or notion crm. – product onboarding → use logic nodes to detect user type (beginner vs. pro) and serve different walkthroughs automatically. – customer support → route basic questions to a support agent, complex ones to human reps, and have it all sync to slack or intercom. it’s still early. the ui has limits, the logic feels clunky, and customization can’t yet match purpose-built platforms like lindy, which give you deeper control and long-term memory. but it's worth taking it for a spin. will be interesting to see how it evolves. and amirmxt does a remarkable job clearly explaining how to build with agentkit etc on The Startup Ideas Podcast (SIP) 🧃 (you can follow for more) full ep on yt will be in the replies. What do you think of OpenAI new agent builder?

Mastering OpenAI's NEW AI AGENT builder in 26 mins: – start with one simple flow (lead capture or faq bot) before adding logic complexity. – use vector stores sparingly - too much context slows performance. – name your nodes clearly so you can debug fast. – integrate with chatkit early if you plan to embed it on a site. – test edge cases (misclassifications, loops, context drops). – think in outcomes, not features - what task can an agent take off your plate today? – treat this as a sandbox, not production - learn what’s possible before scaling. ideas on what you can build (to get your creative juices flowing): – lead qualification → build a multi-agent flow that classifies visitors as hot, warm, or cold, then sends data straight into hubspot or notion crm. – product onboarding → use logic nodes to detect user type (beginner vs. pro) and serve different walkthroughs automatically. – customer support → route basic questions to a support agent, complex ones to human reps, and have it all sync to slack or intercom. it’s still early. the ui has limits, the logic feels clunky, and customization can’t yet match purpose-built platforms like lindy, which give you deeper control and long-term memory. but it's worth taking it for a spin. will be interesting to see how it evolves. and amirmxt does a remarkable job clearly explaining how to build with agentkit etc on The Startup Ideas Podcast (SIP) 🧃 (you can follow for more) full ep on yt will be in the replies. What do you think of OpenAI new agent builder?

GREG ISENBERG

97,315 Aufrufe • vor 10 Monaten

THIS GUY BUILT AN AUTONOMOUS AI AGENT OUT OF CLAUDE CODE + OBSIDIAN and this is way more interesting than another “use AI to take notes” demo the trick is simple: Obsidian is not the writing app here. it becomes the agent’s memory, task board, and context folder. Claude Code is not just answering prompts. it reads the vault, edits files, follows instructions, and keeps moving through the work like a junior operator with a filesystem. the reusable setup looks like this: 1. create an Obsidian vault for one project 2. keep goals, rules, tasks, decisions, and references as markdown files 3. point Claude Code at the folder 4. give it a clear operating loop: read context → choose next task → execute → write back what changed 5. use the notes as persistent memory instead of re-explaining the project every chat that’s the part people miss. the “agent” is not magic. it’s the boring combination of: - local files - explicit rules - task state - write access - a model that can run through the repo/vault Obsidian makes the memory human-readable. Claude Code makes the memory executable. that combo is why the video worked: it turns a notes app into an operating surface for actual work. best use cases: - content systems - research vaults - coding projects - client ops docs - personal knowledge bases that need actions, not just storage the caveat: if your vault is messy, your agent becomes messy too. folders, naming, “done” criteria, and forbidden actions matter more than the prompt. but once the structure is clean, this is one of the easiest ways to build an agent that remembers what happened yesterday without paying for a full custom app.

THIS GUY BUILT AN AUTONOMOUS AI AGENT OUT OF CLAUDE CODE + OBSIDIAN and this is way more interesting than another “use AI to take notes” demo the trick is simple: Obsidian is not the writing app here. it becomes the agent’s memory, task board, and context folder. Claude Code is not just answering prompts. it reads the vault, edits files, follows instructions, and keeps moving through the work like a junior operator with a filesystem. the reusable setup looks like this: 1. create an Obsidian vault for one project 2. keep goals, rules, tasks, decisions, and references as markdown files 3. point Claude Code at the folder 4. give it a clear operating loop: read context → choose next task → execute → write back what changed 5. use the notes as persistent memory instead of re-explaining the project every chat that’s the part people miss. the “agent” is not magic. it’s the boring combination of: - local files - explicit rules - task state - write access - a model that can run through the repo/vault Obsidian makes the memory human-readable. Claude Code makes the memory executable. that combo is why the video worked: it turns a notes app into an operating surface for actual work. best use cases: - content systems - research vaults - coding projects - client ops docs - personal knowledge bases that need actions, not just storage the caveat: if your vault is messy, your agent becomes messy too. folders, naming, “done” criteria, and forbidden actions matter more than the prompt. but once the structure is clean, this is one of the easiest ways to build an agent that remembers what happened yesterday without paying for a full custom app.

kocer

30,403 Aufrufe • vor 2 Monaten

THIS GUY TURNED ONE OBSIDIAN VAULT INTO A SHARED BRAIN FOR 5 CLAUDE AGENTS BUILT TO HANDLE 80% OF A SOLO DEVELOPER’S WORK. 00:03 he zooms out and hundreds of connected notes appear on screen. it is the perfect visual for Claude working from one shared project memory instead of starting from zero every session. most developers give Claude 1 file and 1 prompt. this setup gives it the architecture, coding rules, tests, client briefs, and previous decisions before it writes anything. obsidian becomes the memory layer, storing everything in markdown files Claude can read alongside the codebase. every agent works from the same context instead of repeating the same questions. then 5 specialists split the job. one researches, one plans, one codes, one reviews, and one deploys while the developer only controls the important decisions. one vault, 5 agents, 2 model tiers, and a $50 to $150 monthly stack can turn Claude from a chatbot into a small development team.

THIS GUY TURNED ONE OBSIDIAN VAULT INTO A SHARED BRAIN FOR 5 CLAUDE AGENTS BUILT TO HANDLE 80% OF A SOLO DEVELOPER’S WORK. 00:03 he zooms out and hundreds of connected notes appear on screen. it is the perfect visual for Claude working from one shared project memory instead of starting from zero every session. most developers give Claude 1 file and 1 prompt. this setup gives it the architecture, coding rules, tests, client briefs, and previous decisions before it writes anything. obsidian becomes the memory layer, storing everything in markdown files Claude can read alongside the codebase. every agent works from the same context instead of repeating the same questions. then 5 specialists split the job. one researches, one plans, one codes, one reviews, and one deploys while the developer only controls the important decisions. one vault, 5 agents, 2 model tiers, and a $50 to $150 monthly stack can turn Claude from a chatbot into a small development team.

Gipp 🦅

10,799 Aufrufe • vor 1 Monat

HERMES AGENT HAS 5 SYSTEMS RUNNING UNDER THE HOOD. UNDERSTAND THEM AND YOU USE THE AGENT 10X BETTER. In this video Alejandro AO 🤗 explained: 1. THE AGENT LOOP every message triggers the same cycle: → you send a message → Hermes builds context (SOUL.md + memory.md + user.md + skills + tools + message history) → sends everything to the LLM → LLM decides: call a tool or respond → if tool call: execute, return result, loop back → if response: deliver to you → after response: memory update (agent checks if anything is worth remembering, writes to memory.md or user.md) this loop is why Hermes gets better over time. the memory update after every response means the agent learns from every conversation. 2. CONTEXT ASSEMBLY what the LLM sees on every turn: → SOUL.md (your agent's personality and rules) → memory.md (facts the agent learned over time) → user.md (facts about you, auto-updated) → AGENTS.md and .hermes.md (project context files) → skill descriptions (loaded on demand) → tool schemas (available actions) → message history (current conversation) if SOUL.md is empty, Hermes falls back to a default system prompt. write your own SOUL.md and the agent becomes yours, not generic. CONTEXT COMPRESSION: conversations hit context limits. Hermes handles this at two checkpoints: preflight: before each turn. if conversation exceeds 50% of context window, compression fires. older messages get summarized. last 20 messages stay intact (protect_last_n). gateway auto-compression: between turns. fires at 85%. more aggressive. prevents API errors before the agent even starts processing your message. after compression, a new session lineage ID is generated. the agent can trace back to the original conversation through SQLite. three things break prompt cache: switching models mid-session, changing memory files, or changing context files. 3. THE GATEWAY the system that keeps Hermes reachable on 27+ messaging platforms. an async loop runs continuously. listens for incoming messages from Telegram, Discord, Slack, WhatsApp, email, SMS, and every other adapter. when a message arrives: → gateway identifies which session it belongs to → queries SQLite for the full message history (session ID = platform prefix + chat ID) → builds the context from scratch → sends everything into the agent loop → delivers the response back to the platform the gateway also runs the session manager. when you send a message while the agent is busy: → default: queued for next turn → /steer: injected without interrupting → /interrupt: stops current work without the gateway, Hermes is a CLI tool. with the gateway, Hermes is an always-on agent you reach from your phone. 4. MEMORY (THREE LAYERS) LAYER 1 — MARKDOWN FILES SOUL.md (identity), memory.md (learned facts), user.md (facts about you). injected into context after the system prompt. updated by the agent after every response. LAYER 2 — SQLITE full transcripts of every session stored locally. FTS5 full-text search across all past conversations. session lineage tracking across compressions. the agent can recall what you discussed weeks ago using /recall or session search. LAYER 3 — EXTERNAL PROVIDERS (optional) 8 supported providers: Mem0, SuperMemory, Honcho, Zep, and more. each works differently (semantic search, LLM extraction, similarity matching). queried after the first message in each session. the agent processes your topic first, then checks external memory for related context from past conversations. not enabled by default. enable for significantly better long-term recall. 5. CRON ENGINE a loop inside the gateway ticks every 60 seconds. each tick checks ~/.hermes/cron/jobs.json for scheduled tasks. if a job is due: → fresh session (no chat history, no memory pollution) → execute the prompt with assigned tools → store the run output as markdown in ~/.hermes/cron/output/[job-id]/ → deliver result to your home messaging platform cron does NOT use the send_message tool. delivery happens at the system level, not the agent level. a cron session cannot create more cron jobs. prevents runaway loops. WHY THIS MATTERS: the agent loop teaches it. the context assembly focuses it. the gateway reaches it. the memory remembers it. the cron engine automates it. five systems. one agent. understanding how they connect changes how you configure every level. full 15 levels breakdown in the article 👇

HERMES AGENT HAS 5 SYSTEMS RUNNING UNDER THE HOOD. UNDERSTAND THEM AND YOU USE THE AGENT 10X BETTER. In this video Alejandro AO 🤗 explained: 1. THE AGENT LOOP every message triggers the same cycle: → you send a message → Hermes builds context (SOUL.md + memory.md + user.md + skills + tools + message history) → sends everything to the LLM → LLM decides: call a tool or respond → if tool call: execute, return result, loop back → if response: deliver to you → after response: memory update (agent checks if anything is worth remembering, writes to memory.md or user.md) this loop is why Hermes gets better over time. the memory update after every response means the agent learns from every conversation. 2. CONTEXT ASSEMBLY what the LLM sees on every turn: → SOUL.md (your agent's personality and rules) → memory.md (facts the agent learned over time) → user.md (facts about you, auto-updated) → AGENTS.md and .hermes.md (project context files) → skill descriptions (loaded on demand) → tool schemas (available actions) → message history (current conversation) if SOUL.md is empty, Hermes falls back to a default system prompt. write your own SOUL.md and the agent becomes yours, not generic. CONTEXT COMPRESSION: conversations hit context limits. Hermes handles this at two checkpoints: preflight: before each turn. if conversation exceeds 50% of context window, compression fires. older messages get summarized. last 20 messages stay intact (protect_last_n). gateway auto-compression: between turns. fires at 85%. more aggressive. prevents API errors before the agent even starts processing your message. after compression, a new session lineage ID is generated. the agent can trace back to the original conversation through SQLite. three things break prompt cache: switching models mid-session, changing memory files, or changing context files. 3. THE GATEWAY the system that keeps Hermes reachable on 27+ messaging platforms. an async loop runs continuously. listens for incoming messages from Telegram, Discord, Slack, WhatsApp, email, SMS, and every other adapter. when a message arrives: → gateway identifies which session it belongs to → queries SQLite for the full message history (session ID = platform prefix + chat ID) → builds the context from scratch → sends everything into the agent loop → delivers the response back to the platform the gateway also runs the session manager. when you send a message while the agent is busy: → default: queued for next turn → /steer: injected without interrupting → /interrupt: stops current work without the gateway, Hermes is a CLI tool. with the gateway, Hermes is an always-on agent you reach from your phone. 4. MEMORY (THREE LAYERS) LAYER 1 — MARKDOWN FILES SOUL.md (identity), memory.md (learned facts), user.md (facts about you). injected into context after the system prompt. updated by the agent after every response. LAYER 2 — SQLITE full transcripts of every session stored locally. FTS5 full-text search across all past conversations. session lineage tracking across compressions. the agent can recall what you discussed weeks ago using /recall or session search. LAYER 3 — EXTERNAL PROVIDERS (optional) 8 supported providers: Mem0, SuperMemory, Honcho, Zep, and more. each works differently (semantic search, LLM extraction, similarity matching). queried after the first message in each session. the agent processes your topic first, then checks external memory for related context from past conversations. not enabled by default. enable for significantly better long-term recall. 5. CRON ENGINE a loop inside the gateway ticks every 60 seconds. each tick checks ~/.hermes/cron/jobs.json for scheduled tasks. if a job is due: → fresh session (no chat history, no memory pollution) → execute the prompt with assigned tools → store the run output as markdown in ~/.hermes/cron/output/[job-id]/ → deliver result to your home messaging platform cron does NOT use the send_message tool. delivery happens at the system level, not the agent level. a cron session cannot create more cron jobs. prevents runaway loops. WHY THIS MATTERS: the agent loop teaches it. the context assembly focuses it. the gateway reaches it. the memory remembers it. the cron engine automates it. five systems. one agent. understanding how they connect changes how you configure every level. full 15 levels breakdown in the article 👇

YanXbt

51,860 Aufrufe • vor 2 Monaten

Loved this 22-minute talk on continual learning for AI agents. Must watch for anyone looking to get agents performant and into production. Credit: Soheil Feizi at AI Engineer • Agent learning can happen at three layers: the model (weights), the harness (prompts, tools, skills, code, workflows), and memory (session or persistent). • Two fundamental challenges: (1) getting feedback, meaning how do we know if the agent did well and what it should have done instead, and (2) acting on that feedback, meaning deciding which layer or component to change and how. • Feedback sources differ by stage: In development you have benchmarks with evaluators that score pass/fail. In production you only have logs, which can be judged either automatically (LLMs or code analyzing the log, which is scalable) or by human experts (low volume but critical domain knowledge). • Logs plus feedback aren't enough because they're not testable: A single log with feedback is one observation of what happened. You need to lift it into a replayable learning environment, a simulation with tools, users, and defined evaluators, so candidate fixes can be run, verified, and compared. • Three ways to optimize the agent, with tradeoffs: Model-layer updates (SFT, RL post-training like DPO/GRPO, LoRA) are expensive and need benchmarks and evaluators. Harness updates (trace-to-harness coding agents, prompt search like GEPA) are flexible but either untestable and "vibe-based" or benchmark-dependent. Memory updates (fact storage like Letta/Mem0, skill distillation) are cheapest and fastest but usually unverified. • A good learning engine makes "the smallest durable change at the right layer" of the agent. • Verifiable continual learning (VCL): Improve an agent from its own experience where every fix is proven to help and proven to break nothing that already worked. It requires an executable test (replayable failure), a measured delta (score before and after), and regression tests (prior tests still pass). • Four principles of practical VCL: Replayability (turn one-off failures into rerunnable tests), holisticness (one failure can have causes in memory, prompts, tools, workflow, or model, so route the fix to the right layer), lifelongness (fix new failures subject to no regression on past environments, with regression handled inside the optimization loop rather than post-hoc), and efficiency (the loop must run frequently and cheaply, without scaling linearly as past environments accumulate). • Three takeaways: (1) Agent continual learning isn't necessarily fine-tuning; many useful updates live in the harness and memory layers. (2) Production logs are not learning environments and must be transformed into replayable ones. (3) The frontier is regression-aware improvement: fixing new failures while verifying you don't break old ones.

Loved this 22-minute talk on continual learning for AI agents. Must watch for anyone looking to get agents performant and into production. Credit: Soheil Feizi at AI Engineer • Agent learning can happen at three layers: the model (weights), the harness (prompts, tools, skills, code, workflows), and memory (session or persistent). • Two fundamental challenges: (1) getting feedback, meaning how do we know if the agent did well and what it should have done instead, and (2) acting on that feedback, meaning deciding which layer or component to change and how. • Feedback sources differ by stage: In development you have benchmarks with evaluators that score pass/fail. In production you only have logs, which can be judged either automatically (LLMs or code analyzing the log, which is scalable) or by human experts (low volume but critical domain knowledge). • Logs plus feedback aren't enough because they're not testable: A single log with feedback is one observation of what happened. You need to lift it into a replayable learning environment, a simulation with tools, users, and defined evaluators, so candidate fixes can be run, verified, and compared. • Three ways to optimize the agent, with tradeoffs: Model-layer updates (SFT, RL post-training like DPO/GRPO, LoRA) are expensive and need benchmarks and evaluators. Harness updates (trace-to-harness coding agents, prompt search like GEPA) are flexible but either untestable and "vibe-based" or benchmark-dependent. Memory updates (fact storage like Letta/Mem0, skill distillation) are cheapest and fastest but usually unverified. • A good learning engine makes "the smallest durable change at the right layer" of the agent. • Verifiable continual learning (VCL): Improve an agent from its own experience where every fix is proven to help and proven to break nothing that already worked. It requires an executable test (replayable failure), a measured delta (score before and after), and regression tests (prior tests still pass). • Four principles of practical VCL: Replayability (turn one-off failures into rerunnable tests), holisticness (one failure can have causes in memory, prompts, tools, workflow, or model, so route the fix to the right layer), lifelongness (fix new failures subject to no regression on past environments, with regression handled inside the optimization loop rather than post-hoc), and efficiency (the loop must run frequently and cheaply, without scaling linearly as past environments accumulate). • Three takeaways: (1) Agent continual learning isn't necessarily fine-tuning; many useful updates live in the harness and memory layers. (2) Production logs are not learning environments and must be transformed into replayable ones. (3) The frontier is regression-aware improvement: fixing new failures while verifying you don't break old ones.

Alex Lieberman

20,417 Aufrufe • vor 1 Monat