Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

gpt-6 astra arrived with experimental codex features which are f**king insane... including a new way to carry context across long tasks. [here is what to setup today] 1) context management → notes and searchable history across context windows rollout 2)budgets → track a configured token budget during a task... 3) sleep prevention → keep your computer awake while codex works astra’s experimental context management is off by default. the other switches are worth checking in your installed version too. paste this into codex: “inspect my installed codex version and experimental feature flags. check whether astra’s new context management is fully enabled. explain which other features would help my workflow, verify the exact configuration, and show me a proposed diff before changing anything.” or start with codex features list yourself. a model upgrade is worth exploring beyond the model picker.show more

Avid

25,142 subscribers

47,987 views • 23 hours ago •via X (Twitter)

Science & Technology Education Gaming

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

With all the buzz around the Codex App, OpenAI Developers quietly snuck out a new CLI update (0.94.0) as well. And boy is it an important update! Codex Plan mode is now officially released to the general audience! I am very excited about this one as it has a really strong prompt that is unlike any other plan mode I've personally used. Codex Plan mode doesn't necessarily just ask you 3 questions up front. It goes, collects context, asks questions, collects more context, asks more questions (sometimes), and then writes an incredibly high quality plan. It is my favorite implementation of plan mode thus far. It also comes with Codex's own version of "AskUserQuestion!" Although, it only works in Plan mode for now. They really need to allow people to use it in Code mode as well, but one win at a time. npm i -g @openai/codex Below is a demo of how it works. Let me know what you think!

With all the buzz around the Codex App, OpenAI Developers quietly snuck out a new CLI update (0.94.0) as well. And boy is it an important update! Codex Plan mode is now officially released to the general audience! I am very excited about this one as it has a really strong prompt that is unlike any other plan mode I've personally used. Codex Plan mode doesn't necessarily just ask you 3 questions up front. It goes, collects context, asks questions, collects more context, asks more questions (sometimes), and then writes an incredibly high quality plan. It is my favorite implementation of plan mode thus far. It also comes with Codex's own version of "AskUserQuestion!" Although, it only works in Plan mode for now. They really need to allow people to use it in Code mode as well, but one win at a time. npm i -g @openai/codex Below is a demo of how it works. Let me know what you think!

am.will

57,121 views • 7 months ago

When I talk to serious software engineers, they tell me everyone is sleeping on Codex + GPT 5.2 from OpenAI Which is why I was hyped to get an insiders "how to" from Alexander Embiricos product lead for Codex, showing us how he and the team use Codex to ship almost everything at oai. He'll show us how to: - jump into a new repo and start contributing, even as a non-technical PM - run git worktrees to parallelize agent tasks across codex instances - use best practices and to spec out your work and let codex rip We also talk model vs. harness and how to spy on codex development (which is all OSS!) to understand how to optimize your own implementations. This episode is brought to you by our amazing sponsors 🧠🤑 Brex - The intelligent finance platform built for founders 🦎🐛 Greptile - Your AI code review platform Watch it here:

When I talk to serious software engineers, they tell me everyone is sleeping on Codex + GPT 5.2 from OpenAI Which is why I was hyped to get an insiders "how to" from Alexander Embiricos product lead for Codex, showing us how he and the team use Codex to ship almost everything at oai. He'll show us how to: - jump into a new repo and start contributing, even as a non-technical PM - run git worktrees to parallelize agent tasks across codex instances - use best practices and to spec out your work and let codex rip We also talk model vs. harness and how to spy on codex development (which is all OSS!) to understand how to optimize your own implementations. This episode is brought to you by our amazing sponsors 🧠🤑 Brex - The intelligent finance platform built for founders 🦎🐛 Greptile - Your AI code review platform Watch it here:

claire vo 🖤

48,102 views • 7 months ago

This is crazy, I can talk to OpenAI's realtime API, and it'll respond back with voice, and then help call Codex to help me code! Got curious about Codex and its realtime integration, and found codex's feature flag `realtime_conversation` integrated for the codex app-server. How it works: - launch codex app-server with realtime - connect a client to realtime through WebRTC voice connection - Codex app-server is also attached on the side - talk to realtime - realtime decides whether coding is needed, and responds back with events - Codex app-server takes these events and runs codex agent turns - codex writes code like your codex cli exp Keep in mind OpenAI hasn't released this officially, they're still actively working on it, I just created a fork. This is a completely new UX than the usual chat interface for agentic coding. For example, when the agent responds back in voice, it's actually faster for me to read the response, and I don't actually find voice response that useful unless it does so in the background. But image realtime working in the background, you can talk to it, realtime keeps track of the different agents working on your tasks and your status. More to come on demo side, i'll make a more polished version and include voice, but thought this was cool to share. You can see in the responses the distinction between my voice input, realtime response, and realtime calling Codex to do work!

This is crazy, I can talk to OpenAI's realtime API, and it'll respond back with voice, and then help call Codex to help me code! Got curious about Codex and its realtime integration, and found codex's feature flag `realtime_conversation` integrated for the codex app-server. How it works: - launch codex app-server with realtime - connect a client to realtime through WebRTC voice connection - Codex app-server is also attached on the side - talk to realtime - realtime decides whether coding is needed, and responds back with events - Codex app-server takes these events and runs codex agent turns - codex writes code like your codex cli exp Keep in mind OpenAI hasn't released this officially, they're still actively working on it, I just created a fork. This is a completely new UX than the usual chat interface for agentic coding. For example, when the agent responds back in voice, it's actually faster for me to read the response, and I don't actually find voice response that useful unless it does so in the background. But image realtime working in the background, you can talk to it, realtime keeps track of the different agents working on your tasks and your status. More to come on demo side, i'll make a more polished version and include voice, but thought this was cool to share. You can see in the responses the distinction between my voice input, realtime response, and realtime calling Codex to do work!

Matthew Lam

17,674 views • 3 months ago

i merged all my AI agents into one brain Claude Code, Codex, Gemini: one team, one memory, zero re-briefing result: 8+ hours a week back, and every agent already knows my projects, my clients and how i think here's the setup with Jarvix: 1. dump your context once: projects, clients, writing style, how you make decisions.. it builds your personal model from your actual AI interactions 2. switch agents mid-task with zero re-briefing: Claude Code builds, Codex reviews, Gemini researches, all reading the same context 3. it carries what YOU choose across sessions.. never start a chat from zero again 4. when it's time to act, it pulls the right agents together as one team instead of five separate assistants the unlock isn't another agent. it's that your context finally travels agents without shared memory are five interns on their first day, forever. agents with it are a team that's worked with you for years not more agents. one shared context that's the right bet P.S. thanks to the Jarvix team for the early access

i merged all my AI agents into one brain Claude Code, Codex, Gemini: one team, one memory, zero re-briefing result: 8+ hours a week back, and every agent already knows my projects, my clients and how i think here's the setup with Jarvix: 1. dump your context once: projects, clients, writing style, how you make decisions.. it builds your personal model from your actual AI interactions 2. switch agents mid-task with zero re-briefing: Claude Code builds, Codex reviews, Gemini researches, all reading the same context 3. it carries what YOU choose across sessions.. never start a chat from zero again 4. when it's time to act, it pulls the right agents together as one team instead of five separate assistants the unlock isn't another agent. it's that your context finally travels agents without shared memory are five interns on their first day, forever. agents with it are a team that's worked with you for years not more agents. one shared context that's the right bet P.S. thanks to the Jarvix team for the early access

Ronin

15,071 views • 24 days ago

Introducing Lynk, a brand new way to interact with your favorite harnesses on the go. Compatible with OpenClaw, Hermes, Codex, and local edge models. Fully featured client allowing you to easily and quickly switch between your favorite agents. Lynk has been my absolute favorite way to kick off tasks on the go, completely replacing Telegram. Only available as a beta on Android at present, but iOS version is in development. Works via local network or Tailscale. Features: - Create and continue threads with your favorite harness - OpenClaw, Hermes, Codex, and local models - Android Phone control - Realtime voice agent - Speech-to-text transcription - Codex Pets + live notifications - Draw over screen quick access chat overlay - Open Source software Join the beta in the comments.

am.will

111,376 views • 3 months ago

🚀 Global Launch: Jarvix, Your Context OS is now live! Meet the next generation of personalized AI built around your context, your taste, and your way of thinking. #JarvixOS transforms how you work with AI by understanding your patterns, carrying your context across agents, and helping you move from ideas to action. Step into a new era of AI collaboration: • Discover your AI-native archetype • Build your personal AI model from your evolving AI interactions • Never start from zero - your context moves with you across sessions • Coordinate Codex, Claude Code, Gemini, Hermes, OpenClaw, and more as one connected AI team. • Turn insights into your next move, with you always in control Instead of adapting yourself to every agent, Jarvix carries the context you choose, helps you understand what it reveals, and brings the right agents together when you decide to act. Your AI tools shouldn’t work like separate assistants. They should work like one team. That’s the future of AI: Not more agents. One shared context. One Jarvix.

🚀 Global Launch: Jarvix, Your Context OS is now live! Meet the next generation of personalized AI built around your context, your taste, and your way of thinking. #JarvixOS transforms how you work with AI by understanding your patterns, carrying your context across agents, and helping you move from ideas to action. Step into a new era of AI collaboration: • Discover your AI-native archetype • Build your personal AI model from your evolving AI interactions • Never start from zero - your context moves with you across sessions • Coordinate Codex, Claude Code, Gemini, Hermes, OpenClaw, and more as one connected AI team. • Turn insights into your next move, with you always in control Instead of adapting yourself to every agent, Jarvix carries the context you choose, helps you understand what it reveals, and brings the right agents together when you decide to act. Your AI tools shouldn’t work like separate assistants. They should work like one team. That’s the future of AI: Not more agents. One shared context. One Jarvix.

Jarvix

1,688,727 views • 24 days ago

EVERYTHING YOU NEED TO KNOW ABOUT CHATGPT'S "LOVABLE KILLER" CODEX SITES (in 25 mins): TLDR; the coolest part is that apps you build can update themselves autonomously 1. Codex Sites is not Replit or Lovable or Bolt. Those are great for one-prompting a full app. Codex Sites is for building apps that the agent keeps improving without you touching them. 2. Your personal website can update its own stats. Your internal dashboard can refresh its own data. Your product can add features while you sleep. The app is alive. 3. Start by invoking at-sites. Use realistic sample data. Always say "save for review, do not deploy." This unlocks building a real product, not a homepage. 4. Add persistent storage so the app remembers everything between visits. Without this it resets every time. Ask Codex to show you the data model before it builds. 5. Create safe actions. These are the specific things the agent is allowed to do to your app: add data, update cards, move things, score things. You define the boundaries. The agent operates within them. 6. Build skills so any future Codex chat knows how to interact with your app. The skill is basically a manual for the agent. Without it, every new chat starts from zero. 7. Save gate like a video game. Codex doesn't auto-save. Create checkpoints before you deploy so you can roll back if something breaks. 8. Close the autonomous loop. This is the magic. Once memory, safe actions, and skills are set up, the agent can update your app from any chat, any context, without you switching tabs. 9. Use the plugins most people are sleeping on. Figma, Canva, HeyGen for avatar videos, Game Studio for interactive experiences, FAL for image generation, Hugging Face for open source models. Worth adding a few. 10. The big picture: we went from building apps to raising apps. You set up the structure, the guardrails, and the skills. The agent does the rest. That's autonomous product building and it's here right now. Tbh, Codex sites isn't perfect. Still a lot to be desired like domains, db, authentication etc. But it's a glimpse into this idea that apps can be updated/improved upon automonously. And Codex Sites is REALLY good if you live in Codex everyday. Which more and more of are. And that's really cool. Will be interesting to see how Lovable, Bolt, Replit etc react to this. full tutorial on The Startup Ideas Podcast (SIP) 🧃 where you get your pods watch share with a friend i'm rooting for you What do you think of Codex and Codex sites?

GREG ISENBERG

68,972 views • 3 months ago

Gemini-1.5 Pro has its spotlight stolen today, and people are poking fun at Sora vs Google memes. Well, I think it's the biggest boost in LLM capability so far in 2024. v1.5's 10M token context (1) excels at retrieval; (2) generalizes zero-shot to extremely long instructions like full tutorials and codebases; and (3) works across modalities such as text, audio, and video. Here's a stunning example: v1.5 learns to translate from English to Kalamang purely in context, following a full linguistic manual at inference time. Kalamang is a language spoken by fewer than 200 speakers in western New Guinea. Gemini has never seen this language during training and is only provided with 500 pages of linguistic documentation, a dictionary, and ~400 parallel sentences in context. It basically acquires a sophisticated new skill in the neural activations, instead of gradient finetuning. I talked about the Myth of Context Length many times before: don't get too excited by claims of 1M or even 1B context tokens. LSTMs already achieved literally infinite context length 25 yrs ago! What truly matters is how well the model actually uses the context to solve real-world problems, and Gemini-1.5 has surpassed the SOTA with flying colors. The paper is also well-written with lots of solid quantitative analysis on in-context memorization and generalization. Paper: “Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context” Congrats to Jeff Dean Oriol Vinyals Sundar Pichai and team!

Gemini-1.5 Pro has its spotlight stolen today, and people are poking fun at Sora vs Google memes. Well, I think it's the biggest boost in LLM capability so far in 2024. v1.5's 10M token context (1) excels at retrieval; (2) generalizes zero-shot to extremely long instructions like full tutorials and codebases; and (3) works across modalities such as text, audio, and video. Here's a stunning example: v1.5 learns to translate from English to Kalamang purely in context, following a full linguistic manual at inference time. Kalamang is a language spoken by fewer than 200 speakers in western New Guinea. Gemini has never seen this language during training and is only provided with 500 pages of linguistic documentation, a dictionary, and ~400 parallel sentences in context. It basically acquires a sophisticated new skill in the neural activations, instead of gradient finetuning. I talked about the Myth of Context Length many times before: don't get too excited by claims of 1M or even 1B context tokens. LSTMs already achieved literally infinite context length 25 yrs ago! What truly matters is how well the model actually uses the context to solve real-world problems, and Gemini-1.5 has surpassed the SOTA with flying colors. The paper is also well-written with lots of solid quantitative analysis on in-context memorization and generalization. Paper: “Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context” Congrats to Jeff Dean Oriol Vinyals Sundar Pichai and team!

Jim Fan

278,517 views • 2 years ago

Everyone wants agent swarms. Very few people are talking seriously enough about the context layer that makes swarms useful. Even with one agent, context is fragile. Too little context and the agent guesses. Too much context and it wastes tokens, loses focus, or reasons over irrelevant noise. The sweet spot is precise context: the right knowledge, in the right structure, at the right moment. With many agents, that challenge explodes. Each agent produces decisions, assumptions, findings, summaries, risks, and partial conclusions. Unless that knowledge becomes shared, structured, and reusable, every new agent is forced to rediscover what another agent already learned. That is not a swarm. That is a crowd. Shared context graphs are what turn agent activity into agent collaboration, and OriginTrail DKG V10 brings them to life. Was just playing with some final polishing for the V10 release, and it is really powerful to see shared context graphs where multiple agents contribute knowledge into the same connected memory, with attribution visible directly in the graph ui. That matters for three reasons. First, agents can access and build on one shared memory instead of staying trapped in isolated sessions. Second, the graph structure helps them retrieve the exact context they need, instead of stuffing everything into a prompt and hoping the model sorts it out. Third, verifiability of provenance. You can see which agent contributed each piece of knowledge, trace the source, and decide what to trust. Tokenmaxxing starts with fewer tokens, but the deeper story is coordination - agents stop reloading the world and start building on shared, verifiable context. That is the foundation for serious multi-agent work across software engineering, research, finance, operations, project management, and far beyond. The future is not more agents, it is agents working from shared, verifiable context. But the more the merrier, of course.

Everyone wants agent swarms. Very few people are talking seriously enough about the context layer that makes swarms useful. Even with one agent, context is fragile. Too little context and the agent guesses. Too much context and it wastes tokens, loses focus, or reasons over irrelevant noise. The sweet spot is precise context: the right knowledge, in the right structure, at the right moment. With many agents, that challenge explodes. Each agent produces decisions, assumptions, findings, summaries, risks, and partial conclusions. Unless that knowledge becomes shared, structured, and reusable, every new agent is forced to rediscover what another agent already learned. That is not a swarm. That is a crowd. Shared context graphs are what turn agent activity into agent collaboration, and OriginTrail DKG V10 brings them to life. Was just playing with some final polishing for the V10 release, and it is really powerful to see shared context graphs where multiple agents contribute knowledge into the same connected memory, with attribution visible directly in the graph ui. That matters for three reasons. First, agents can access and build on one shared memory instead of staying trapped in isolated sessions. Second, the graph structure helps them retrieve the exact context they need, instead of stuffing everything into a prompt and hoping the model sorts it out. Third, verifiability of provenance. You can see which agent contributed each piece of knowledge, trace the source, and decide what to trust. Tokenmaxxing starts with fewer tokens, but the deeper story is coordination - agents stop reloading the world and start building on shared, verifiable context. That is the foundation for serious multi-agent work across software engineering, research, finance, operations, project management, and far beyond. The future is not more agents, it is agents working from shared, verifiable context. But the more the merrier, of course.

Jurij Skornik

11,180 views • 3 months ago

New short course: LLMs as Operating Systems: Agent Memory, created with Letta, and taught by its founders Charles Packer and Sarah Wooders. An LLM's input context window has limited space. Using a longer input context also costs more and results in slower processing. So, managing what's stored in this context window is important. In the innovative paper MemGPT: Towards LLMs as Operating Systems, its authors (which include the instructors) proposed using an LLM agent to manage this context window. Their system uses a large persistent memory that stores everything that could be included in the input context, and an agent decides what is actually included. Take the example of building a chatbot that needs to remember what's been said earlier in a conversation (perhaps over many days of interaction with a user). As the conversation's length grows, the memory management agent will move information from the input context to a persistent searchable database; summarize information to keep relevant facts in the input context; and restore relevant conversation elements from further back in time. This allows a chatbot to keep what's currently most relevant in its input context memory to generate the next response. When I read the original MemGPT paper, I thought it was an innovative technique for handling memory for LLMs. The open-source Letta framework, which we'll use in this course, makes MemGPT easy to implement. It adds memory to your LLM agents and gives them transparent long-term memory. In detail, you’ll learn: - How to build an agent that can edit its own limited input context memory, using tools and multi-step reasoning - What is a memory hierarchy (an idea from computer operating systems, which use a cache to speed up memory access), and how these ideas apply to managing the LLM input context (where the input context window is a "cache" storing the most relevant information; and an agent decides what to move in and out of this to/from a larger persistent storage system) - How to implement multi-agent collaboration by letting different agents share blocks of memory This course will give you a sophisticated understanding of memory management for LLMs, which is important for chatbots having long conversations, and for complex agentic workflows. Please sign up here!

New short course: LLMs as Operating Systems: Agent Memory, created with Letta, and taught by its founders Charles Packer and Sarah Wooders. An LLM's input context window has limited space. Using a longer input context also costs more and results in slower processing. So, managing what's stored in this context window is important. In the innovative paper MemGPT: Towards LLMs as Operating Systems, its authors (which include the instructors) proposed using an LLM agent to manage this context window. Their system uses a large persistent memory that stores everything that could be included in the input context, and an agent decides what is actually included. Take the example of building a chatbot that needs to remember what's been said earlier in a conversation (perhaps over many days of interaction with a user). As the conversation's length grows, the memory management agent will move information from the input context to a persistent searchable database; summarize information to keep relevant facts in the input context; and restore relevant conversation elements from further back in time. This allows a chatbot to keep what's currently most relevant in its input context memory to generate the next response. When I read the original MemGPT paper, I thought it was an innovative technique for handling memory for LLMs. The open-source Letta framework, which we'll use in this course, makes MemGPT easy to implement. It adds memory to your LLM agents and gives them transparent long-term memory. In detail, you’ll learn: - How to build an agent that can edit its own limited input context memory, using tools and multi-step reasoning - What is a memory hierarchy (an idea from computer operating systems, which use a cache to speed up memory access), and how these ideas apply to managing the LLM input context (where the input context window is a "cache" storing the most relevant information; and an agent decides what to move in and out of this to/from a larger persistent storage system) - How to implement multi-agent collaboration by letting different agents share blocks of memory This course will give you a sophisticated understanding of memory management for LLMs, which is important for chatbots having long conversations, and for complex agentic workflows. Please sign up here!

Andrew Ng

201,127 views • 1 year ago

How can you solve complex tasks using a Large Language Model? Here is a 2-minute introduction to everything you need to know to 10x the quality of your results. Let's talk about three techniques, in order of complexity, starting with the easiest one: • In-Context Learning • Indexing + In-Context Learning • Fine-tuning In-Context Learning The team that trained GPT-3 found something they couldn't explain: You can condition a model using examples of how you want it to behave. I included an example prompt in the attached video. You can "teach" the model how you want it to interpret questions, select the correct answers, and format the results by giving a few examples. You can also give specific knowledge to the model that will be helpful when formulating answers. We call this approach "grounding the model." There's another example in the video. Indexing + In-Context Learning Unfortunately, there is a limit to how much data you can include in a prompt. We call this the "context size." One version of GPT-4 supports a context of approximately 6,000 words, while the other supports 25,000 words. Although this sounds like a lot, many applications need more than that. Imagine you wrote a book and want to build an application to answer any questions about your story. What happens if your book is longer than the context? That's where Indexing comes in. Using a model, you can turn every book passage into an embedding. These are vectors, numbers that "encode" the passage's text. You can then store these embeddings in a particular database that supports fast retrieval of these vectors. You can then turn any question into an embedding and search the database for the list of passages that are similar to that query. Instead of using the entire book to ask the model, you can now use the relevant passages as in-context information, effectively working around the context size limitation. Fine-tuning Fine-tuning can give you an extra boost to get reliable outputs from your LLM. It is, however, the most complex approach on the list. There are different approaches to fine-tuning a model with your data. A popular technique is to process your data with your LLM and use the outputs to train a new classifier that solves your specific task. Notice that here you aren't modifying the LLM. Instead, you are chaining it with your trained classifier. Another approach is to modify the parameters of the LLM using your data. Think of this as "rewiring" the model in a way that solves your particular task. The results and costs will vary depending on how many layers you want to fine-tune from the original model. Many companies think that fine-tuning is the solution to their problems. In my experience, many will benefit from exploring the other two approaches. I love explaining Machine Learning and Artificial Intelligence ideas. If you enjoy in-depth content like this, follow me Santiago so you don't miss what comes next.

How can you solve complex tasks using a Large Language Model? Here is a 2-minute introduction to everything you need to know to 10x the quality of your results. Let's talk about three techniques, in order of complexity, starting with the easiest one: • In-Context Learning • Indexing + In-Context Learning • Fine-tuning In-Context Learning The team that trained GPT-3 found something they couldn't explain: You can condition a model using examples of how you want it to behave. I included an example prompt in the attached video. You can "teach" the model how you want it to interpret questions, select the correct answers, and format the results by giving a few examples. You can also give specific knowledge to the model that will be helpful when formulating answers. We call this approach "grounding the model." There's another example in the video. Indexing + In-Context Learning Unfortunately, there is a limit to how much data you can include in a prompt. We call this the "context size." One version of GPT-4 supports a context of approximately 6,000 words, while the other supports 25,000 words. Although this sounds like a lot, many applications need more than that. Imagine you wrote a book and want to build an application to answer any questions about your story. What happens if your book is longer than the context? That's where Indexing comes in. Using a model, you can turn every book passage into an embedding. These are vectors, numbers that "encode" the passage's text. You can then store these embeddings in a particular database that supports fast retrieval of these vectors. You can then turn any question into an embedding and search the database for the list of passages that are similar to that query. Instead of using the entire book to ask the model, you can now use the relevant passages as in-context information, effectively working around the context size limitation. Fine-tuning Fine-tuning can give you an extra boost to get reliable outputs from your LLM. It is, however, the most complex approach on the list. There are different approaches to fine-tuning a model with your data. A popular technique is to process your data with your LLM and use the outputs to train a new classifier that solves your specific task. Notice that here you aren't modifying the LLM. Instead, you are chaining it with your trained classifier. Another approach is to modify the parameters of the LLM using your data. Think of this as "rewiring" the model in a way that solves your particular task. The results and costs will vary depending on how many layers you want to fine-tune from the original model. Many companies think that fine-tuning is the solution to their problems. In my experience, many will benefit from exploring the other two approaches. I love explaining Machine Learning and Artificial Intelligence ideas. If you enjoy in-depth content like this, follow me Santiago so you don't miss what comes next.

Santiago

384,510 views • 3 years ago

I just added real-time AI translation into Chormex using GPT-Realtime-2… and this feels absolutely surreal. It works across YouTube videos, live streams, meetings, presentations, basically anywhere audio is playing inside Chrome. You can watch translated speech in real time while simultaneously using Codex on top of the live context. “Summarize this.” “What are the key points?” “Turn this into notes.” “Explain what they mean.” “Organize the discussion.” …all while the video or meeting is still happening. It genuinely feels like browsers are evolving into real-time AI operating systems. We are getting dangerously close to a world where language barriers on the internet completely disappear.

I just added real-time AI translation into Chormex using GPT-Realtime-2… and this feels absolutely surreal. It works across YouTube videos, live streams, meetings, presentations, basically anywhere audio is playing inside Chrome. You can watch translated speech in real time while simultaneously using Codex on top of the live context. “Summarize this.” “What are the key points?” “Turn this into notes.” “Explain what they mean.” “Organize the discussion.” …all while the video or meeting is still happening. It genuinely feels like browsers are evolving into real-time AI operating systems. We are getting dangerously close to a world where language barriers on the internet completely disappear.

CHOI

192,215 views • 4 months ago

here is the first look at realtime voice mode in codex the ui is already in the app, but the codex backend endpoint does not seem to be live yet i got it running by forcing the local feature gate and using my own openai api key, so it goes through the public realtime api instead it's incredible, but this demo alone cost me ~$4 gpt realtime 2 is very expensive, which makes me wonder how they're going to handle it - part of the normal usage limit, or a separate limit like we had with gpt 5.3 codex spark? i'm turning it off until the official rollout but get excited, the next few weeks are going to be amazing

here is the first look at realtime voice mode in codex the ui is already in the app, but the codex backend endpoint does not seem to be live yet i got it running by forcing the local feature gate and using my own openai api key, so it goes through the public realtime api instead it's incredible, but this demo alone cost me ~$4 gpt realtime 2 is very expensive, which makes me wonder how they're going to handle it - part of the normal usage limit, or a separate limit like we had with gpt 5.3 codex spark? i'm turning it off until the official rollout but get excited, the next few weeks are going to be amazing

🥔🥔🥔

39,508 views • 3 months ago

New course: Spec-Driven Development with Coding Agents, built in partnership with JetBrains, and taught by Paul Everitt | @pauleveritt@fosstodon.org. Vibe coding is fast, but often produces code that doesn't match what you asked for. This short course teaches you spec-driven development: write a detailed spec defining what to build, and work with your coding agent to implement it. Many of the best developers already build this way. A spec lets you control large code changes with a few words, preserve context across agent sessions, and stay in control as your project grows in complexity. Skills you'll gain: - Write a detailed specification to define your mission, tech stack, and roadmap, giving your agent the context it needs from the start - Plan, implement, and validate features in iterative loops using a spec as your agent's guide - Apply the same repeatable workflow to both new and legacy codebases - Package your workflow into a portable agent skill that works across agents and IDEs Join and write specs that keep your coding agent on track!

New course: Spec-Driven Development with Coding Agents, built in partnership with JetBrains, and taught by Paul Everitt | @[email protected]. Vibe coding is fast, but often produces code that doesn't match what you asked for. This short course teaches you spec-driven development: write a detailed spec defining what to build, and work with your coding agent to implement it. Many of the best developers already build this way. A spec lets you control large code changes with a few words, preserve context across agent sessions, and stay in control as your project grows in complexity. Skills you'll gain: - Write a detailed specification to define your mission, tech stack, and roadmap, giving your agent the context it needs from the start - Plan, implement, and validate features in iterative loops using a spec as your agent's guide - Apply the same repeatable workflow to both new and legacy codebases - Package your workflow into a portable agent skill that works across agents and IDEs Join and write specs that keep your coding agent on track!

Andrew Ng

464,592 views • 4 months ago

Buzz is Slack where the agents are teammates instead of integrations. Here's what that unlocks: Agents: - The ones you start with are just defaults - Edit them, add your own, write their instructions like a system prompt Swappable Harness: - This is the killer feature - Swap Claude Code, Codex, Goose under the same agent Context: - Every chat you've had with that agent carries over to the new harness - Global skills and skill files come too Think: Slack = the shape Agents = the teammates Harness = the engine you can change without losing anything "I get model fatigue" is a real thing. Buzz sits one layer above it.

Buzz is Slack where the agents are teammates instead of integrations. Here's what that unlocks: Agents: - The ones you start with are just defaults - Edit them, add your own, write their instructions like a system prompt Swappable Harness: - This is the killer feature - Swap Claude Code, Codex, Goose under the same agent Context: - Every chat you've had with that agent carries over to the new harness - Global skills and skill files come too Think: Slack = the shape Agents = the teammates Harness = the engine you can change without losing anything "I get model fatigue" is a real thing. Buzz sits one layer above it.

The Startup Ideas Podcast (SIP) 🧃

43,523 views • 1 month ago

The highest leverage thing you have to optimize for AI agents is their context. If you’re not getting enough out of AI agents, you probably have a context problem. By default the AI model knows absolutely nothing relevant to perform the task successfully. The whole job of context engineering is to prioritize what the model has as information to work with. Small deviations in the goals, instructions, domain understanding, or underlying data you supply the model will lead to wildly different outcomes. This is where you should spend almost all your time and energy in a workflow. We will start to religiously document more of our workflows and business processes, create clearer goals for people and agents, and ensure our tech infra is setup to get AI agents the right information. The lessons are currently being learned the fastest with coding agents, but the same lessons will apply to every field of knowledge work going forward. dex has a great talk on this that’s absolutely worth your time to watch.

The highest leverage thing you have to optimize for AI agents is their context. If you’re not getting enough out of AI agents, you probably have a context problem. By default the AI model knows absolutely nothing relevant to perform the task successfully. The whole job of context engineering is to prioritize what the model has as information to work with. Small deviations in the goals, instructions, domain understanding, or underlying data you supply the model will lead to wildly different outcomes. This is where you should spend almost all your time and energy in a workflow. We will start to religiously document more of our workflows and business processes, create clearer goals for people and agents, and ensure our tech infra is setup to get AI agents the right information. The lessons are currently being learned the fastest with coding agents, but the same lessons will apply to every field of knowledge work going forward. dex has a great talk on this that’s absolutely worth your time to watch.

Aaron Levie

87,086 views • 1 year ago

Loop engineering is great until something breaks. Here is how I improve the reliability of my agentic loops. I use human-in-the-loop (HITL). It's easy and extremely effective. Anyone can build this. My setup: I recorded a quick demo of how it all works. I shared recently that I now use more voice agents to build and communicate with agents. I also use them to verify. I hate the idea of being tied down to my computer or in a Slack channel to communicate with my agents. Here is what I have done to streamline communication with my agents. All my Claude and Codex agent sessions now use the Dial MCP server. It has a bunch of tools and provisions my agents with their own number that can place calls as native tools, with voice, SMS, and iMessage behind one interface. As my loops/automations work on PRs and new features, my agents escalate decisions to me via a short phone call. This is extremely useful when I am on the road or away from my desk. If you want to try this with Claude Code or Codex, paste this into your agent and get started right away: "Get yourself a Dial phone number and call me. Say hello and that setup is working, then hang up. Follow Natan Voitenkov and team are building something special here. Go check them out. Give your agent a phone number now: ($5 free credit)

Loop engineering is great until something breaks. Here is how I improve the reliability of my agentic loops. I use human-in-the-loop (HITL). It's easy and extremely effective. Anyone can build this. My setup: I recorded a quick demo of how it all works. I shared recently that I now use more voice agents to build and communicate with agents. I also use them to verify. I hate the idea of being tied down to my computer or in a Slack channel to communicate with my agents. Here is what I have done to streamline communication with my agents. All my Claude and Codex agent sessions now use the Dial MCP server. It has a bunch of tools and provisions my agents with their own number that can place calls as native tools, with voice, SMS, and iMessage behind one interface. As my loops/automations work on PRs and new features, my agents escalate decisions to me via a short phone call. This is extremely useful when I am on the road or away from my desk. If you want to try this with Claude Code or Codex, paste this into your agent and get started right away: "Get yourself a Dial phone number and call me. Say hello and that setup is working, then hang up. Follow Natan Voitenkov and team are building something special here. Go check them out. Give your agent a phone number now: ($5 free credit)

elvis

15,295 views • 2 months ago

.OpenAI's GPT-5.1 is live in Cline. After testing the last several weeks, here's what we've found makes GPT-5.1 an exceptional model in Cline: GPT-5.1 is an obsessive researcher. It reads everything before writing anything. Where other models grab context and go, GPT-5.1 investigates thoroughly first, even for one-line changes. We found this overkill for simple tasks, and modified GPT-5.1's prompting resultantly, but a superpower when channeled right. Here's what you should expect from GPT-5.1 in Cline: > deep, systematic investigation before any edits > follows constraints religiously (specs, todo lists, your requirements) > maintains coherent reasoning with large contexts > delivers complete features with fewer corrections How we optimized Cline for GPT-5.1: 1. redesigned prompts for execution focus and stricter Plan/Act transitions 2. tightened tool specs, GPT-5.1 internalizes rules with just single examples 3. split /deep-planning investigation into two phases (read first, then search) Result: GPT-5.1 is a model that finishes what it starts, even across dozens of files and thousands of lines. Both GPT-5.1 (272K context) and GPT-5.1-codex (400K context) available now in Cline. $1.25/M input, $10/M output. Enable native tool calling in settings for best results. Use /deep-planning to see GPT-5.1 at its best. Full technical breakdown linked below.

.OpenAI's GPT-5.1 is live in Cline. After testing the last several weeks, here's what we've found makes GPT-5.1 an exceptional model in Cline: GPT-5.1 is an obsessive researcher. It reads everything before writing anything. Where other models grab context and go, GPT-5.1 investigates thoroughly first, even for one-line changes. We found this overkill for simple tasks, and modified GPT-5.1's prompting resultantly, but a superpower when channeled right. Here's what you should expect from GPT-5.1 in Cline: > deep, systematic investigation before any edits > follows constraints religiously (specs, todo lists, your requirements) > maintains coherent reasoning with large contexts > delivers complete features with fewer corrections How we optimized Cline for GPT-5.1: 1. redesigned prompts for execution focus and stricter Plan/Act transitions 2. tightened tool specs, GPT-5.1 internalizes rules with just single examples 3. split /deep-planning investigation into two phases (read first, then search) Result: GPT-5.1 is a model that finishes what it starts, even across dozens of files and thousands of lines. Both GPT-5.1 (272K context) and GPT-5.1-codex (400K context) available now in Cline. $1.25/M input, $10/M output. Enable native tool calling in settings for best results. Use /deep-planning to see GPT-5.1 at its best. Full technical breakdown linked below.

Cline

72,111 views • 9 months ago

Three months ago, Codex was trash for knowledge work. Now it's my daily driver. I use it for writing, recruiting, deep engineering work, and everything in between. It even keeps me at inbox 0. I chatted with Every 🧱's head of growth Austin Austin Tedesco on Every 🧱's AI & I about what changed, and why he now spends 80% of his working time in the Codex desktop app too. We get into: - How Codex went from making Austin feel like an idiot to being the place he goes to get stuff done, including complex tasks like writing go-to-market plans using existing material from Slack, Notion, and meeting transcripts. - Why the Codex’s desktop app, which is faster and more reliable than Claude Desktop/Cowork, is the real differentiator. - How I source candidates with Codex by having it identify career arcs, not keywords—my go-to move is identifying organizations likely to teach the skills Every needs for a role, and then find candidates from that pool who have since gone on to work in AI. This is a must-watch for anyone who's wondering whether it’s finally time to give Codex a try. Watch below! Timestamps How Codex went from a tool for senior engineers to a daily driver for knowledge work: 00:00:57 How Claude Code proved that a great coding agent works for any knowledge work: 00:02:42 Austin's switch to Codex: 00:07:24 How Austin set up Codex with folders, keys, and reviewer agents: 00:13:48 Using Codex to brainstorm automations across Gmail, Slack, and Notion: 00:18:24 How Austin manages the human review step when Codex is drafting communications: 00:22:42 Using Codex to build specialized agents inspired by product executive Claire Vo: 00:28:54 Synthesizing meeting transcripts and Slack threads into a go-to-market plan: 00:31:09 Building a live KPI tracker in Notion that agents can read: 00:40:15 Using Codex for recruiting: 00:44:54

Three months ago, Codex was trash for knowledge work. Now it's my daily driver. I use it for writing, recruiting, deep engineering work, and everything in between. It even keeps me at inbox 0. I chatted with Every 🧱's head of growth Austin Austin Tedesco on Every 🧱's AI & I about what changed, and why he now spends 80% of his working time in the Codex desktop app too. We get into: - How Codex went from making Austin feel like an idiot to being the place he goes to get stuff done, including complex tasks like writing go-to-market plans using existing material from Slack, Notion, and meeting transcripts. - Why the Codex’s desktop app, which is faster and more reliable than Claude Desktop/Cowork, is the real differentiator. - How I source candidates with Codex by having it identify career arcs, not keywords—my go-to move is identifying organizations likely to teach the skills Every needs for a role, and then find candidates from that pool who have since gone on to work in AI. This is a must-watch for anyone who's wondering whether it’s finally time to give Codex a try. Watch below! Timestamps How Codex went from a tool for senior engineers to a daily driver for knowledge work: 00:00:57 How Claude Code proved that a great coding agent works for any knowledge work: 00:02:42 Austin's switch to Codex: 00:07:24 How Austin set up Codex with folders, keys, and reviewer agents: 00:13:48 Using Codex to brainstorm automations across Gmail, Slack, and Notion: 00:18:24 How Austin manages the human review step when Codex is drafting communications: 00:22:42 Using Codex to build specialized agents inspired by product executive Claire Vo: 00:28:54 Synthesizing meeting transcripts and Slack threads into a go-to-market plan: 00:31:09 Building a live KPI tracker in Notion that agents can read: 00:40:15 Using Codex for recruiting: 00:44:54

Dan Shipper 📧

55,561 views • 4 months ago

$HERMES AGENT CAN SHARE MEMORY WITH CODEX AND CLAUDE CODE THROUGH HINDSIGHT. ONE MEMORY BANK. ONE AGENT REMEMBERS, EVERY OTHER AGENT KNOWS. the problem: you use Hermes for orchestration. Codex for coding. Claude Code for debugging. each has its own memory. switch between them and you explain the same project three times. Hindsight fixes this. one shared memory bank that every agent reads and writes to. tell Codex: "the test color for this project is purple." switch to Hermes. ask: "what test color did I pick?" Hermes answers: "purple." no copy-paste. no re-explaining. instant recall. HOW IT WORKS: Hindsight runs as a Docker container on your machine. self-hosted. your data stays local. an LLM powers the memory processing (retain, recall, reflect). RETAIN: extracts facts from your conversations. entities, decisions, preferences, project context. saved to the memory bank automatically. RECALL: when you ask a question, Hindsight pulls from semantic search, keywords, graph connections, and temporal data. fused into one answer. REFLECT: deeper reasoning layer. connects memories across sessions. identifies patterns in your work. produces observations that get smarter over time. CONNECT TO HERMES: Desktop app: Settings → Memory and Context → switch provider from Namosin to Hindsight. set API URL to your local Docker container. set bank ID. done. CLI: hermes memory setup → select Hindsight. verify: hermes memory status should show: provider: hindsight, installed, available. CONNECT TO CODEX: npx hindsight-coding-agents install codex \ --self-hosted --server this installs lifecycle hooks: initialize memory on session start. recall context during work. retain the session when done. enable hooks in Codex: Settings → Hooks → trust all three. CONNECT TO CLAUDE CODE (same command): npx hindsight-coding-agents install all "all" connects every detected agent on your machine. Claude Code, Codex, Cursor, and others. one command. every agent shares the same bank. TAGS FOR FILTERING: every memory gets tagged by harness (Hermes, Codex, Claude Code) and optionally by project name. in the Hindsight control plane: filter by harness. see only Hermes memories. or only Codex memories. or search across everything. soft partitions inside one bank. not hard walls. cross-reference when you need to. ONE BANK OR MANY: one global bank: solo dev, related projects. all agents share everything. patterns emerge across projects. per-project banks: unrelated codebases. each project gets its own memory. no cross-contamination. your call. start with one. split when projects diverge. KNOWLEDGE PAGES (v0.9.0): Hindsight auto-generates living summaries from your accumulated memories. components, concepts, conventions, decisions. not static docs. projected from real agent conversations. auto-refresh as new memories land. WHAT TO KNOW: self-hosted via Docker. your data never leaves your machine. backup system built in (admin CLI + scheduled exports). works with any LLM (local Ollama, OpenAI, Codex subscription). memory defense: redact or block sensitive content automatically. 33,000+ memories accumulated in ~10 days of normal use.$

HERMES AGENT CAN SHARE MEMORY WITH CODEX AND CLAUDE CODE THROUGH HINDSIGHT. ONE MEMORY BANK. ONE AGENT REMEMBERS, EVERY OTHER AGENT KNOWS. the problem: you use Hermes for orchestration. Codex for coding. Claude Code for debugging. each has its own memory. switch between them and you explain the same project three times. Hindsight fixes this. one shared memory bank that every agent reads and writes to. tell Codex: "the test color for this project is purple." switch to Hermes. ask: "what test color did I pick?" Hermes answers: "purple." no copy-paste. no re-explaining. instant recall. HOW IT WORKS: Hindsight runs as a Docker container on your machine. self-hosted. your data stays local. an LLM powers the memory processing (retain, recall, reflect). RETAIN: extracts facts from your conversations. entities, decisions, preferences, project context. saved to the memory bank automatically. RECALL: when you ask a question, Hindsight pulls from semantic search, keywords, graph connections, and temporal data. fused into one answer. REFLECT: deeper reasoning layer. connects memories across sessions. identifies patterns in your work. produces observations that get smarter over time. CONNECT TO HERMES: Desktop app: Settings → Memory and Context → switch provider from Namosin to Hindsight. set API URL to your local Docker container. set bank ID. done. CLI: hermes memory setup → select Hindsight. verify: hermes memory status should show: provider: hindsight, installed, available. CONNECT TO CODEX: npx hindsight-coding-agents install codex \ --self-hosted --server this installs lifecycle hooks: initialize memory on session start. recall context during work. retain the session when done. enable hooks in Codex: Settings → Hooks → trust all three. CONNECT TO CLAUDE CODE (same command): npx hindsight-coding-agents install all "all" connects every detected agent on your machine. Claude Code, Codex, Cursor, and others. one command. every agent shares the same bank. TAGS FOR FILTERING: every memory gets tagged by harness (Hermes, Codex, Claude Code) and optionally by project name. in the Hindsight control plane: filter by harness. see only Hermes memories. or only Codex memories. or search across everything. soft partitions inside one bank. not hard walls. cross-reference when you need to. ONE BANK OR MANY: one global bank: solo dev, related projects. all agents share everything. patterns emerge across projects. per-project banks: unrelated codebases. each project gets its own memory. no cross-contamination. your call. start with one. split when projects diverge. KNOWLEDGE PAGES (v0.9.0): Hindsight auto-generates living summaries from your accumulated memories. components, concepts, conventions, decisions. not static docs. projected from real agent conversations. auto-refresh as new memories land. WHAT TO KNOW: self-hosted via Docker. your data never leaves your machine. backup system built in (admin CLI + scheduled exports). works with any LLM (local Ollama, OpenAI, Codex subscription). memory defense: redact or block sensitive content automatically. 33,000+ memories accumulated in ~10 days of normal use.

YanXbt

29,200 views • 18 days ago