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.

Finally, an OpenRouter for agent harnesses! (including System One by Jev) Devs just open-sourced a plug-and-play infrastructure layer that lets you run any harness under a single interface, like: - Codex - Hermes - Claude code - DeepSeek Harness - System One, powered by Jev - And 9 more... agent harnesses This means you can bring Jev into the same product that already uses Codex, Claude Code, or another supported harness, without writing another implementation for sessions, streaming, files, cancellation, and failure handling. Here's the repo: (don't forget to star it ⭐ ) The harnesses run locally, and the Unified Harness Protocol (UHP) defines the common task interface with an OpenAI Responses-compatible API. If you want to dive deeper, my recent article explains why model routing is not the same as harness routing, and what it takes to support multiple harnesses. It also covers UHP, the full local setup, a working API call, and how sessions and files work. Read it below.show more

Akshay 🚀

288,865 subscribers

170,457 views • 20 hours ago •via X (Twitter)

Education Science & Technology

Anya Rossi• Live Now

Private livecam show

34 Comments

Jatin Garg19 hours ago

OpenRouter works because all the models behind it take the same shape — text in, text out. Agent harnesses have different state, tool signatures, and session models. How does a unified interface handle that?

Akmal Makhmudov20 hours ago

I only run one harness for my own build, Claude Code, so I've never hit the switching-cost pain this solves directly. for someone already committed to one harness, is the real win here A/B testing models across harnesses, or something else?

Yev Mura17 hours ago

Useful layer. Separate itch on my side: once Codex / Claude Code / etc are installed they still rubber-stamp themselves. I built consilium so the hub agent can `consult` a different-provider CLI for a second opinion, or run a council that verifies every file:line finding.

Labangram Photography19 hours ago

The hard part seems to be semantic parity: session resume, streamed tool events, and cancellation need identical guarantees across harnesses. A conformance suite covering those cases would make UHP much more useful than an adapter layer.

TechP20 hours ago

This could make building with multiple AI agents much simpler. One interface, multiple harnesses.

Posterior19 hours ago

What would be really cool is if the router automatically chooses the correct harness for a given task.

Kero Dewto20 hours ago

Interesting! I’ll try it out for mobile development and then share my hands-on experience on my page

Kizuno1820 hours ago

standardizing the harness layer is the missing piece. in our autonomous pipeline targeting brazil, we don't bind models to tasks directly. the harness runs deterministic python stages and invokes external harnesses only when schema gates fail. harness routing > model routing

Hussain Hashim | Building SundayBack17 hours ago

@akshay_pachaar that's a game-changer for devs dealing with multiple harnesses! simplifies the workflow like crazy. really makes you think about how much time we waste on integrations.

Chris Izatt19 hours ago

Yoooo lemme finish up TaskWraith so it qualifies 🥲

BTCaveman.18 hours ago

OM >

catman18 hours ago

The unified interface is the strongest part, but harness routing is not quite equivalent to model routing: harnesses carry different state, tool semantics, and failure modes. UHP can normalize the plumbing, not erase those behavioral differences.

cit1z3n17 hours ago

routing models is easy. routing harnesses is the part people pay for twice.

shaun19 hours ago

one harness already needs a babysitter. now i can route between five lol

Salise18 hours ago

the idea of running multiple harnesses like Codex and Claude under one interface is really cool, should save a lot of time

Zam16 hours ago

A shared harness interface is useful only if it standardizes the ugly parts too, especially cancellation, state handoff, and failure recovery across sessions.

saietta19 hours ago

model routing works because the contract is thin -- messages in, tokens out. harness routing means normalizing side effects: how each one sandboxes shell access, what a cancellation leaves mid-file-write, when a permission prompt blocks vs auto-approves. that's where it leaks first.

EDDY VU18 hours ago

Having a clean abstraction layer to benchmark different harnesses against the exact same task suite is huge.

ShadowAguy17 hours ago

Unified interface sounds useful, but wondering how you handle the different prompt/response formats across Codex, Claude Code, and DeepSeek? Does the abstraction leak when one tool expects JSON while another wants raw text?

Later17 hours ago

Curious how you measured harness-to-harness overhead — same task, same model, what’s the extra latency and tokens just from the unified protocol versus talking to Claude Code or Codex directly?

DeDi18 hours ago

专为代理框架设计，UHP统一会话流式取消这类脏活实现

Later18 hours ago

Unifying the harness interface is useful, but did anyone run the same task set across Codex / Claude Code / Jev and report tokens per solved task plus idle turns?

kuzzo18 hours ago

We already have @omnigent_ai, you’re overlooking this…

Arbaz17 hours ago

harness switcher is nice. finding one that barely drains is better

techa16 hours ago

the agent wars are just starting and standardizing the harness layer is already table stakes, whoever owns that interface wins.

Soni19 hours ago

Standardizing permissions and sessions sounds clean until two runtimes disagree on the schema. Still babysitting, just one level higher up.

Alex | RouteAPI18 hours ago

A common protocol gets you portability. State and tool behavior decide whether the switch is actually painless. Are you using more than one coding harness today?

Adam · Arcyton18 hours ago

One interface across harnesses is useful mostly for comparison. The part worth normalising along with the calls is failure: each harness reports a failed tool step differently, and some report it as a normal answer. Does the layer unify errors, or pass them through?

Sophia Data Queen17 hours ago

One interface for the harness, not just the model.

Gregor19 hours ago

not sure the OpenRouter analogy holds. OpenRouter works because models share a shape. harness behavior is the opinionated part, and a wrapper flattens exactly that. does it preserve Jev's context-carry or does that disappear?

Saeed Ahmed17 hours ago

Under one harness interface, what breaks first when you swap Codex for Claude Code: permissions, tools, or memory?

JSCC | Nasdaq · AI17 hours ago

Unified Harness Protocol can list Codex. Sessions and cancellation still have to work across every harness.

Sebastiano Mandalà18 hours ago

I honestly start to be fed up about these hype driven, AI written tools and articles that add complexity instead of value.

Max Bevza18 hours ago

this is actually massive for agent infra

Related Videos

"The model is no longer the product. Codex, Perplexity Computer, or Claude Code - all are orchestration system. It takes a model and pairs it with an agent harness. What is an agent harness ? The rules for how the agent loops around" - Aravind Srinivas

"The model is no longer the product. Codex, Perplexity Computer, or Claude Code - all are orchestration system. It takes a model and pairs it with an agent harness. What is an agent harness ? The rules for how the agent loops around" - Aravind Srinivas

Rohan Paul

120,488 views • 3 months ago

Finally, an open-source runtime security layer for your agent harnesses. AI agents can call tools, run shell commands, modify files, and interact with production systems. But when something goes wrong, teams often have to reconstruct what the agent actually did from scattered logs after the fact. Agent Beacon is built to change that. It runs locally and records agent activity as it happens, including: → Tool calls → Shell commands → File changes → Approval decisions → Session and harness context The useful part is that it normalizes all of this into the same event format across 23+ agent harnesses. So instead of writing different detection logic for Claude Code, Codex, or another harness, security teams can reason about the underlying action itself. A command executed is a command executed, regardless of which harness produced it. Beacon also records how confidently an event was captured. An action can be directly observed from the runtime or inferred from indirect evidence. That matters when you start writing security rules against this data. For example, you can detect when an agent reads an environment variable, modifies a sensitive file, executes a suspicious command, or performs an action without the expected approval. And the detection happens while the session is still unfolding, not days later during incident reconstruction. Everything runs locally by default. You can inspect sessions through the dashboard, write your own detection rules, or forward the same normalized events to tools like Splunk, Datadog, Elastic, Sentinel, or CrowdStrike. Beacon GitHub: (don't forget to star 🌟) I also wrote a detailed article on the same, covering how important runtime security is and how to build it for your agent harnesses. The article is quoted below.

Finally, an open-source runtime security layer for your agent harnesses. AI agents can call tools, run shell commands, modify files, and interact with production systems. But when something goes wrong, teams often have to reconstruct what the agent actually did from scattered logs after the fact. Agent Beacon is built to change that. It runs locally and records agent activity as it happens, including: → Tool calls → Shell commands → File changes → Approval decisions → Session and harness context The useful part is that it normalizes all of this into the same event format across 23+ agent harnesses. So instead of writing different detection logic for Claude Code, Codex, or another harness, security teams can reason about the underlying action itself. A command executed is a command executed, regardless of which harness produced it. Beacon also records how confidently an event was captured. An action can be directly observed from the runtime or inferred from indirect evidence. That matters when you start writing security rules against this data. For example, you can detect when an agent reads an environment variable, modifies a sensitive file, executes a suspicious command, or performs an action without the expected approval. And the detection happens while the session is still unfolding, not days later during incident reconstruction. Everything runs locally by default. You can inspect sessions through the dashboard, write your own detection rules, or forward the same normalized events to tools like Splunk, Datadog, Elastic, Sentinel, or CrowdStrike. Beacon GitHub: (don't forget to star 🌟) I also wrote a detailed article on the same, covering how important runtime security is and how to build it for your agent harnesses. The article is quoted below.

Akshay 🚀

27,361 views • 11 days ago

run agent harnesses 100% private & offline. (no token costs, no API keys, 100% open-source) your agent runs locally. the model doesn't. every prompt, every file, and every secret still leaves your machine before the agent does anything with it. Magnitude fixes that. it's an open source inference server that runs models on your own hardware and plugs into the coding agent you already use. setup is one command. it profiles your machine, measures the memory bandwidth that sets your token rate, and hands back complete configurations instead of a list of models. each one names a model, a compression level, a context size, and a speed range you can expect. pick one and start working. it doesn't replace your harness. setup asks which one you want and writes that config for you. Pi, OpenCode, Claude Code, Codex, and Cline all work, and there's a built-in one tuned for local models if you don't have a harness yet. that one uses your shell, edits files, and runs scripts out of the box. add skills and it handles Excel, PowerPoint, PDFs, or Chrome. everyday work it covers: → analyze sensitive data → manage private notes → review code and logs → search and organize files → build docs or slides Apache 2.0. no rate limits, and nothing leaves the machine. 𝗻𝗽𝗺 𝗶 -𝗴 @𝗺𝗮𝗴𝗻𝗶𝘁𝘂𝗱𝗲𝗱𝗲𝘃/𝗰𝗹𝗶 the repo is here: (don't forget to star 🌟) i wrote the full breakdown of why picking the configuration is the hard part. the article is quoted below.

run agent harnesses 100% private & offline. (no token costs, no API keys, 100% open-source) your agent runs locally. the model doesn't. every prompt, every file, and every secret still leaves your machine before the agent does anything with it. Magnitude fixes that. it's an open source inference server that runs models on your own hardware and plugs into the coding agent you already use. setup is one command. it profiles your machine, measures the memory bandwidth that sets your token rate, and hands back complete configurations instead of a list of models. each one names a model, a compression level, a context size, and a speed range you can expect. pick one and start working. it doesn't replace your harness. setup asks which one you want and writes that config for you. Pi, OpenCode, Claude Code, Codex, and Cline all work, and there's a built-in one tuned for local models if you don't have a harness yet. that one uses your shell, edits files, and runs scripts out of the box. add skills and it handles Excel, PowerPoint, PDFs, or Chrome. everyday work it covers: → analyze sensitive data → manage private notes → review code and logs → search and organize files → build docs or slides Apache 2.0. no rate limits, and nothing leaves the machine. 𝗻𝗽𝗺 𝗶 -𝗴 @𝗺𝗮𝗴𝗻𝗶𝘁𝘂𝗱𝗲𝗱𝗲𝘃/𝗰𝗹𝗶 the repo is here: (don't forget to star 🌟) i wrote the full breakdown of why picking the configuration is the hard part. the article is quoted below.

Akshay 🚀

55,693 views • 19 days ago

The easiest way to run your agent harnesses using local models: Whether you're using Claude Code, Codex, OpenCode, or Pi, the local setup itself isn't really the hard part. The harder problem is figuring out which local model your machine can actually handle well. You have to think about RAM, model size, quantization, context length, KV cache, speed, accuracy, and a bunch of other trade-offs before you even start. Magnitude is an open-source solution that removes all of that guesswork for you. It profiles your machine, benchmarks what it can realistically run, recommends the best models for your hardware, and then lets you connect them to your preferred agent harness. The entire setup takes just two commands. In this video, I walk through the full process from hardware profiling to running a harness on a local model. Chapters: 00:00 Intro 00:17 What Magnitude is 00:45 Setup in two commands 01:29 Why picking a local model gets confusing fast 02:11 Hardware profiling + model recommendations 02:59 Choosing your harness 03:22 Connecting a harness that's already running 04:18 Outro Get started: (don't forget to star 🌟)

The easiest way to run your agent harnesses using local models: Whether you're using Claude Code, Codex, OpenCode, or Pi, the local setup itself isn't really the hard part. The harder problem is figuring out which local model your machine can actually handle well. You have to think about RAM, model size, quantization, context length, KV cache, speed, accuracy, and a bunch of other trade-offs before you even start. Magnitude is an open-source solution that removes all of that guesswork for you. It profiles your machine, benchmarks what it can realistically run, recommends the best models for your hardware, and then lets you connect them to your preferred agent harness. The entire setup takes just two commands. In this video, I walk through the full process from hardware profiling to running a harness on a local model. Chapters: 00:00 Intro 00:17 What Magnitude is 00:45 Setup in two commands 01:29 Why picking a local model gets confusing fast 02:11 Hardware profiling + model recommendations 02:59 Choosing your harness 03:22 Connecting a harness that's already running 04:18 Outro Get started: (don't forget to star 🌟)

Akshay 🚀

18,532 views • 6 days 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

Andrew Ng just released a free 2-hour course on complete Harness Engineering How to go from one prompt to a reliable system of agents that can run, test, and improve themselves: 09:14 - Build your first agent from scratch 33:11 - Master agent loops 1:02:46 - Turn loops into reliable workflows 1:30:15 - Build agents that improve their own work 1:49:05 - Run the complete system without supervision Model → Harness → Reliable Software Most agent tutorials stop once the model can call a tool This one shows you how to build the infrastructure around it within the first 20 minutes Most people are still prompting one agent at a time Andrew Ng is already teaching the layer above: Harnesses that give agents context, tools, tests, and feedback Watch this brilliant course and build the harness Then read the full architecture below ↓

Andrew Ng just released a free 2-hour course on complete Harness Engineering How to go from one prompt to a reliable system of agents that can run, test, and improve themselves: 09:14 - Build your first agent from scratch 33:11 - Master agent loops 1:02:46 - Turn loops into reliable workflows 1:30:15 - Build agents that improve their own work 1:49:05 - Run the complete system without supervision Model → Harness → Reliable Software Most agent tutorials stop once the model can call a tool This one shows you how to build the infrastructure around it within the first 20 minutes Most people are still prompting one agent at a time Andrew Ng is already teaching the layer above: Harnesses that give agents context, tools, tests, and feedback Watch this brilliant course and build the harness Then read the full architecture below ↓

Ichigo

100,062 views • 17 days ago

Anthropic Technical Staff Jess Yan says the harness and the model cannot be pulled apart without giving up performance, though she flags her own bias. "I think I am quite biased but I also think that it is impossible to get the maximum possible performance without tying together the harness and the model." "Now the components of the harness and maybe the thickness of the harness will change over time as models get more and more capable." "However, when we test our models and when we are assessing their performance we always have to test it in conjunction with a harness." "And are we going to test it with all the different harnesses of the world? We're going to select the harnesses that we have built." "And so there is an aspect of the necessity of building models is that you have to be testing them with harnesses, and that sort of keeps them paired together."

Anthropic Technical Staff Jess Yan says the harness and the model cannot be pulled apart without giving up performance, though she flags her own bias. "I think I am quite biased but I also think that it is impossible to get the maximum possible performance without tying together the harness and the model." "Now the components of the harness and maybe the thickness of the harness will change over time as models get more and more capable." "However, when we test our models and when we are assessing their performance we always have to test it in conjunction with a harness." "And are we going to test it with all the different harnesses of the world? We're going to select the harnesses that we have built." "And so there is an aspect of the necessity of building models is that you have to be testing them with harnesses, and that sort of keeps them paired together."

Fireside Alpha

584,074 views • 1 month ago

introducing matrix. the ultimate multi-agent runtime that turns your product into actual revenue and automates 99% of what it takes to run a company. probably the most different, most future-feeling agent product you've ever seen. the harness, the interface, everything is rebuilt from scratch. think of it as the complement to codex and claude code. they write the code. matrix does literally everything else you actually need done. plug in your codex subscription and run it for free, or byok. fully free either way. this video was edited by matrix and voiced by an agent. more real cases coming soon. matrix is in beta preview right now. full launch next week. retweet + comment and we'll send invite codes to the first 200.

introducing matrix. the ultimate multi-agent runtime that turns your product into actual revenue and automates 99% of what it takes to run a company. probably the most different, most future-feeling agent product you've ever seen. the harness, the interface, everything is rebuilt from scratch. think of it as the complement to codex and claude code. they write the code. matrix does literally everything else you actually need done. plug in your codex subscription and run it for free, or byok. fully free either way. this video was edited by matrix and voiced by an agent. more real cases coming soon. matrix is in beta preview right now. full launch next week. retweet + comment and we'll send invite codes to the first 200.

Derek Nee

15,361 views • 3 months ago

Anthropic engineer: "At Anthropic, we don't write prompts anymore We build harnesses" In just 42 minutes, she shows how the Claude team builds harnesses that prompt themselves Agents → Harness → Loops → Graphs → Self-Improving Systems If this were a $500 course, people would call it one of the best agent courses of the year But it's completely free And it's better than most paid agent engineering courses Watch it today Then use the step-by-step harness engineering guide below

Anthropic engineer: "At Anthropic, we don't write prompts anymore We build harnesses" In just 42 minutes, she shows how the Claude team builds harnesses that prompt themselves Agents → Harness → Loops → Graphs → Self-Improving Systems If this were a $500 course, people would call it one of the best agent courses of the year But it's completely free And it's better than most paid agent engineering courses Watch it today Then use the step-by-step harness engineering guide below

rari

32,547 views • 21 days ago

The founder of LangChain says both models and harnesses have gotten really good between December and now. According to Harrison Chase, the core idea of an agent before Christmas was a model running in a loop and calling tools. This had been the north star for 3 years. - langchain had this when it launched - autogpt was the same idea - openclaw is kind of a future version of it Then about a year ago, they started getting really good. Claude Code, Manus, and Deep Research were all launched around the same time. All of them use the same pattern: running in a loop with harnesses (planning tools, file systems, code execution, etc) Harness engineering became a thing. Then Opus came out in November and really unlocked it. - the harness let the model do more and more - less hardcoded logic - way more control Then everyone went on vacation, played around, and realized that the model and the harness finally worked reliably.

The founder of LangChain says both models and harnesses have gotten really good between December and now. According to Harrison Chase, the core idea of an agent before Christmas was a model running in a loop and calling tools. This had been the north star for 3 years. - langchain had this when it launched - autogpt was the same idea - openclaw is kind of a future version of it Then about a year ago, they started getting really good. Claude Code, Manus, and Deep Research were all launched around the same time. All of them use the same pattern: running in a loop with harnesses (planning tools, file systems, code execution, etc) Harness engineering became a thing. Then Opus came out in November and really unlocked it. - the harness let the model do more and more - less hardcoded logic - way more control Then everyone went on vacation, played around, and realized that the model and the harness finally worked reliably.

Ivan Burazin

33,948 views • 5 months ago

Perplexity CEO Aravind Srinivas on the brutal truth about who actually makes money in AI (and why it's not who you think): Aravind argues that the real value in AI comes from orchestration. He points to products like Codex, Claude Code, and Perplexity Computer: "What is that? It's an orchestration system. It takes a model, pairs it with an agent harness." And what is an agent harness? "The simplest way of describing it is like rules for how the agent loop should run. What are all the skills and sub-agents and connectors and tools it accesses? Without the harness, you don't necessarily capture and convert the intrinsic intelligence in the model into valuable output tokens." This leads to a blunt conclusion about who has a real business in AI, and who doesn't: "If you're literally just a reseller of model tokens, you have no business, because the model will get commoditized. So even if you're a model builder, you don't have a business. As an infra layer, you have some business on serving those output tokens. But as an application layer or model builder, you don't really have a business if you're just a reseller of tokens that come directly out of the model." So where does the value accrue? "You have a business if you know how to take the model, ground it in valuable context, orchestrate it with a really good agent harness, connected to the right set of tools and connectors (whether it's personal connectors or business connectors) and provide the experience to people in one single unified system." Aravind Srinivas then explains Perplexity's specific edge: Beyond orchestrating across tools, files, and connectors, they also orchestrate across models. "That is the differentiation that Anthropic and OpenAI cannot claim, because you wouldn't find GPT-5 inside the Claude Code harness. You wouldn't find Claude Opus inside the Codex harness. These are competing with each other. Whereas you would find both these models inside Perplexity Computer." Why does this matter? Because it all comes down to power. In Aravind's framing, the fundamental cost driver in AI is watts (the one input nobody can subsidize except the government). "Whoever provides the most valuable output tokens with the least amount of power expended to produce them generates the greatest value to the end user, has the most pricing power, has the most value. That is the orchestration problem to solve." His conclusion: "The one single most important metric in AI is token value per watt per user."

Perplexity CEO Aravind Srinivas on the brutal truth about who actually makes money in AI (and why it's not who you think): Aravind argues that the real value in AI comes from orchestration. He points to products like Codex, Claude Code, and Perplexity Computer: "What is that? It's an orchestration system. It takes a model, pairs it with an agent harness." And what is an agent harness? "The simplest way of describing it is like rules for how the agent loop should run. What are all the skills and sub-agents and connectors and tools it accesses? Without the harness, you don't necessarily capture and convert the intrinsic intelligence in the model into valuable output tokens." This leads to a blunt conclusion about who has a real business in AI, and who doesn't: "If you're literally just a reseller of model tokens, you have no business, because the model will get commoditized. So even if you're a model builder, you don't have a business. As an infra layer, you have some business on serving those output tokens. But as an application layer or model builder, you don't really have a business if you're just a reseller of tokens that come directly out of the model." So where does the value accrue? "You have a business if you know how to take the model, ground it in valuable context, orchestrate it with a really good agent harness, connected to the right set of tools and connectors (whether it's personal connectors or business connectors) and provide the experience to people in one single unified system." Aravind Srinivas then explains Perplexity's specific edge: Beyond orchestrating across tools, files, and connectors, they also orchestrate across models. "That is the differentiation that Anthropic and OpenAI cannot claim, because you wouldn't find GPT-5 inside the Claude Code harness. You wouldn't find Claude Opus inside the Codex harness. These are competing with each other. Whereas you would find both these models inside Perplexity Computer." Why does this matter? Because it all comes down to power. In Aravind's framing, the fundamental cost driver in AI is watts (the one input nobody can subsidize except the government). "Whoever provides the most valuable output tokens with the least amount of power expended to produce them generates the greatest value to the end user, has the most pricing power, has the most value. That is the orchestration problem to solve." His conclusion: "The one single most important metric in AI is token value per watt per user."

Big Brain AI

42,484 views • 1 month ago

OpenEnv has a new home: starting today, it's coordinated by a committee that includes Meta-PyTorch, Reflection, Unsloth, Modal, Prime Intellect, Nvidia, Mercor, Fleet AI, and Hugging Face frontier labs train their models and their harnesses together. Claude knows Claude Code. GPT-5.5 knows Codex. that's not an accident, it's training. open-source models deserve the same magic, but pulling that off requires infrastructure that belongs to everyone, not one lab OpenEnv is that layer. one api, any harness, any trainer, any environment rewards and training loops stay in TRL, Unsloth, wherever you already work. OpenEnv is the socket they all plug into get involved!

OpenEnv has a new home: starting today, it's coordinated by a committee that includes Meta-PyTorch, Reflection, Unsloth, Modal, Prime Intellect, Nvidia, Mercor, Fleet AI, and Hugging Face frontier labs train their models and their harnesses together. Claude knows Claude Code. GPT-5.5 knows Codex. that's not an accident, it's training. open-source models deserve the same magic, but pulling that off requires infrastructure that belongs to everyone, not one lab OpenEnv is that layer. one api, any harness, any trainer, any environment rewards and training loops stay in TRL, Unsloth, wherever you already work. OpenEnv is the socket they all plug into get involved!

Sergio Paniego

32,085 views • 3 months ago

Google Brain founder Andrew Ng: "Prompting will be dead in 6 months Agent harnesses built with loops and graphs will replace it" Agent → Harness → Feedback → Loops → Graphs → Self-Improving Systems In this 1-hour Stanford lecture, he explains what the best engineers are building instead and how you can start today Prompt → Run → Verify → Improve The first 20 minutes teach what most $1,500 courses try to sell you For free Bookmark and watch it today Then read the guide below on building an agent harness that improves itself

Google Brain founder Andrew Ng: "Prompting will be dead in 6 months Agent harnesses built with loops and graphs will replace it" Agent → Harness → Feedback → Loops → Graphs → Self-Improving Systems In this 1-hour Stanford lecture, he explains what the best engineers are building instead and how you can start today Prompt → Run → Verify → Improve The first 20 minutes teach what most $1,500 courses try to sell you For free Bookmark and watch it today Then read the guide below on building an agent harness that improves itself

rari

122,176 views • 2 days ago

One of the easiest wins with an AI agent right now is letting it sign up for your own mobile app on a real phone. There's a repo called Phone Harness that connects Codex or Claude Code to an actual iPhone or Android. The agent sees the screen and can tap, type, scroll, open apps, and verify what happened. Here's the whole setup: 1) Connect the phone. iPhone goes through Mac iPhone mirroring, Android goes through ADB over USB or wireless. 2) You don't have to jailbreak anything, you don't need Xcode, and you don't need a special app on the phone. 3) Paste the setup prompt from the repo into Codex or Claude Code: clone the GitHub, read the install.md, put phone harness on your path, register it as an agent skill, then read onboarding.md and walk me through setup. 4) Run phone-harness 5) Give it the first job: open the app, create an account, tap through onboarding, try checkout, take screenshots, and tell me where it gets confusing or breaks. 6) Then point it at the mobile apps that have no API at all. Every mobile team should be running that signup test constantly. Most don't, because it's tedious, or they don't want to hire that person, or the person makes mistakes anyway. Agents are moving from answering questions to operating tools, and the phone is one of the biggest tool surfaces in the world. Connect the phone, run the doctor, then hand it your signup flow

One of the easiest wins with an AI agent right now is letting it sign up for your own mobile app on a real phone. There's a repo called Phone Harness that connects Codex or Claude Code to an actual iPhone or Android. The agent sees the screen and can tap, type, scroll, open apps, and verify what happened. Here's the whole setup: 1) Connect the phone. iPhone goes through Mac iPhone mirroring, Android goes through ADB over USB or wireless. 2) You don't have to jailbreak anything, you don't need Xcode, and you don't need a special app on the phone. 3) Paste the setup prompt from the repo into Codex or Claude Code: clone the GitHub, read the install.md, put phone harness on your path, register it as an agent skill, then read onboarding.md and walk me through setup. 4) Run phone-harness 5) Give it the first job: open the app, create an account, tap through onboarding, try checkout, take screenshots, and tell me where it gets confusing or breaks. 6) Then point it at the mobile apps that have no API at all. Every mobile team should be running that signup test constantly. Most don't, because it's tedious, or they don't want to hire that person, or the person makes mistakes anyway. Agents are moving from answering questions to operating tools, and the phone is one of the biggest tool surfaces in the world. Connect the phone, run the doctor, then hand it your signup flow

The Startup Ideas Podcast (SIP) 🧃

25,561 views • 18 days ago

In the future, you’ll be able to accomplish a goal by just giving Claude an outcome and a budget. That’s the direction Anthropic is building in with its new Managed Agents features, announced at this week’s Code with Claude developer event. The basic idea: Claude, wrapped in a computer in the cloud, that you can spin up, scale, and manage as needed. Anthropic is taking on the infrastructure that kills most agent products, and making sure that it scales to meet the needs of agents running 24/7. On this week’s AI & I from Every 📧, I talk with Angela Jiang (Angela Jiang), head of product for the Claude platform, and Katelyn Lesse (Katelyn Lesse), head of engineering for the Claude platform, about what Anthropic is building and what it takes to make agents reliable in production. We get into: - Why the "build a generic harness, hot-swap any model behind it" playbook is already outdated. Angela points to eval data on Memory where the same task across different harnesses performed drastically differently. - The infrastructure wall every team hits in production—and why Katelyn thinks “my sandbox died and took the agent with it” is the real reason internal agents don't ship. - Why Anthropic is so bullish on using file systems and skills within Claude, including Angela's argument that those early design choices can compound for years. This is a must-watch for anyone trying to take an agent past the demo and into production. Watch below! Timestamps: How the Claude platform evolved from API to agents: 00:01:48 The primitives that make up Claude Managed Agents: 00:04:09 Why the harness and the model are becoming a single unit: 00:10:37 The infrastructure wall that kills most agent projects in production: 00:18:49 Why team agents need a different shape than individual productivity tools: 00:24:49 How Anthropic's legal team uses an agent to review marketing copy: 00:26:36 Using multi-agent orchestration for advisor strategies, adversarial pairs, and swarms: 00:34:24 How to measure agent success with outcome and budget as the end state: 00:35:50 What the platform looks like a year from now, when Claude writes its own harness: 00:39:11

In the future, you’ll be able to accomplish a goal by just giving Claude an outcome and a budget. That’s the direction Anthropic is building in with its new Managed Agents features, announced at this week’s Code with Claude developer event. The basic idea: Claude, wrapped in a computer in the cloud, that you can spin up, scale, and manage as needed. Anthropic is taking on the infrastructure that kills most agent products, and making sure that it scales to meet the needs of agents running 24/7. On this week’s AI & I from Every 📧, I talk with Angela Jiang (Angela Jiang), head of product for the Claude platform, and Katelyn Lesse (Katelyn Lesse), head of engineering for the Claude platform, about what Anthropic is building and what it takes to make agents reliable in production. We get into: - Why the "build a generic harness, hot-swap any model behind it" playbook is already outdated. Angela points to eval data on Memory where the same task across different harnesses performed drastically differently. - The infrastructure wall every team hits in production—and why Katelyn thinks “my sandbox died and took the agent with it” is the real reason internal agents don't ship. - Why Anthropic is so bullish on using file systems and skills within Claude, including Angela's argument that those early design choices can compound for years. This is a must-watch for anyone trying to take an agent past the demo and into production. Watch below! Timestamps: How the Claude platform evolved from API to agents: 00:01:48 The primitives that make up Claude Managed Agents: 00:04:09 Why the harness and the model are becoming a single unit: 00:10:37 The infrastructure wall that kills most agent projects in production: 00:18:49 Why team agents need a different shape than individual productivity tools: 00:24:49 How Anthropic's legal team uses an agent to review marketing copy: 00:26:36 Using multi-agent orchestration for advisor strategies, adversarial pairs, and swarms: 00:34:24 How to measure agent success with outcome and budget as the end state: 00:35:50 What the platform looks like a year from now, when Claude writes its own harness: 00:39:11

Dan Shipper

66,862 views • 4 months ago

Harness vs. Graphs, clearly explained! a harness is great, and most people think it is the whole thing: retries, timeouts, a sandbox, a log, the context it assembles before every call. all of that is real work, and all of it wraps exactly one call. run it a hundred times and you have one call, made very safely, a hundred times. Graph engineering fixes this by moving the decision up a layer: not how safely one call is made, but which calls exist to be made at all. you need both, and here is the sentence that resolves the whole confusion: the harness is everything around one call. the graph is everything between them. ↳ around one call: retry, timeout, sandbox, log, assemble the context, hand back a result ↳ between calls: split, fan out, merge, gate, send back Prompts → Context → Harness → Loops → Graphs the harness does not go away when you build a graph. it moves under each node, and now there are five of them, each wrapping a call you would never have made by hand. the trick is knowing which layer a failure belongs to. turn a piece off and run it again. if the call still works, it was the harness. if the wrong step runs at all, it was the graph. people spend weeks hardening a harness around a node that should not have existed. one thing to know before you scale it. most of what people call their agent is a harness with a chat box on it. ↳ it retries, it times out, it logs, it assembles context, it holds one call up beautifully ↳ it has never once decided that a second call should exist, and that is the entire difference that last one catches careful people. a harness that never fails is not evidence the system is right. it is evidence one call went well, which is the smallest possible claim. and the one that eats whole nights: a harness cannot save you from the wrong step running. you can retry a bad decision three times with a clean log and perfect isolation, and all you bought was three copies of it. below i have quoted my full guide on graph engineering. it covers the three topologies, the verifier patterns, and where the gate should actually open. save this and read it below ↓

Harness vs. Graphs, clearly explained! a harness is great, and most people think it is the whole thing: retries, timeouts, a sandbox, a log, the context it assembles before every call. all of that is real work, and all of it wraps exactly one call. run it a hundred times and you have one call, made very safely, a hundred times. Graph engineering fixes this by moving the decision up a layer: not how safely one call is made, but which calls exist to be made at all. you need both, and here is the sentence that resolves the whole confusion: the harness is everything around one call. the graph is everything between them. ↳ around one call: retry, timeout, sandbox, log, assemble the context, hand back a result ↳ between calls: split, fan out, merge, gate, send back Prompts → Context → Harness → Loops → Graphs the harness does not go away when you build a graph. it moves under each node, and now there are five of them, each wrapping a call you would never have made by hand. the trick is knowing which layer a failure belongs to. turn a piece off and run it again. if the call still works, it was the harness. if the wrong step runs at all, it was the graph. people spend weeks hardening a harness around a node that should not have existed. one thing to know before you scale it. most of what people call their agent is a harness with a chat box on it. ↳ it retries, it times out, it logs, it assembles context, it holds one call up beautifully ↳ it has never once decided that a second call should exist, and that is the entire difference that last one catches careful people. a harness that never fails is not evidence the system is right. it is evidence one call went well, which is the smallest possible claim. and the one that eats whole nights: a harness cannot save you from the wrong step running. you can retry a bad decision three times with a clean log and perfect isolation, and all you bought was three copies of it. below i have quoted my full guide on graph engineering. it covers the three topologies, the verifier patterns, and where the gate should actually open. save this and read it below ↓

Hanako

50,770 views • 10 days ago

Jev Founder, Diogo Almeida (ex-OpenAI): "The next era is not the Claude Code era. Claude Code is still part of the assistance era - Jev is the next step Jev is x200 faster, x400 cheaper and has 0 hallucination. This is what comes after RLHF" In 17 minutes he explains why every model you use has a human-shaped flaw baked into it and why Jev is better worth more than a $500 Claude Code engineering course watch today, then read how to actually boost yourself x100 with Jev in the post below

Jev Founder, Diogo Almeida (ex-OpenAI): "The next era is not the Claude Code era. Claude Code is still part of the assistance era - Jev is the next step Jev is x200 faster, x400 cheaper and has 0 hallucination. This is what comes after RLHF" In 17 minutes he explains why every model you use has a human-shaped flaw baked into it and why Jev is better worth more than a $500 Claude Code engineering course watch today, then read how to actually boost yourself x100 with Jev in the post below

Codez

186,400 views • 3 days ago

There's a built-in agent inside Claude Code you can invoke just by telling Claude to use the claude-code-guide agent Save this... you'll need it It's not a local file in your project. It's a built-in subagent defined in Claude Code's system itself. It answers questions about Claude Code CLI, the Agent SDK, and the Claude API. It uses Glob, Grep, Read, WebFetch, and WebSearch, but can't edit or write files. Its only job is to answer questions.

There's a built-in agent inside Claude Code you can invoke just by telling Claude to use the claude-code-guide agent Save this... you'll need it It's not a local file in your project. It's a built-in subagent defined in Claude Code's system itself. It answers questions about Claude Code CLI, the Agent SDK, and the Claude API. It uses Glob, Grep, Read, WebFetch, and WebSearch, but can't edit or write files. Its only job is to answer questions.

Daniel San

57,445 views • 6 months ago

Google's team published a 9-page PDF on Harness Engineering. one formula replaced prompt engineering: Agent = Model + Harness the twist: same Claude Sonnet, same benchmark. change only the harness the playbook, six steps: step 1 -> add guides AGENTS.md, rule files, constraint docs. every line is a past agent failure turned into a permanent fix step 2 -> add sensors linters, tests, validation scripts the agent runs on its own output before a human sees it step 3 -> build the agentic loop plan, execute, verify, fix. bounded retries, budget caps, escalation when stuck step 4 -> externalize memory the model forgets every session. the harness holds state, decisions and artifacts across all of them step 5 -> enforce permissions which tools, how many writes, what needs approval. safety lives in the harness, never in the model step 6 -> wire observability track every tool call, cost and retry. trip wires fire when behavior drifts the result: your agent turns from a demo into infrastructure every failure makes the system permanently better, not just the next conversation that's also the line between an agent you show people and one a client pays you to leave running one harness gets you a reliable loop. connect several and it becomes graph engineering, where one agent's failure turns into another agent's sensor the PDF stops at the harness. the graph is the next layer that breakdown, running on Kimi, is below ↓

Google's team published a 9-page PDF on Harness Engineering. one formula replaced prompt engineering: Agent = Model + Harness the twist: same Claude Sonnet, same benchmark. change only the harness the playbook, six steps: step 1 -> add guides AGENTS.md, rule files, constraint docs. every line is a past agent failure turned into a permanent fix step 2 -> add sensors linters, tests, validation scripts the agent runs on its own output before a human sees it step 3 -> build the agentic loop plan, execute, verify, fix. bounded retries, budget caps, escalation when stuck step 4 -> externalize memory the model forgets every session. the harness holds state, decisions and artifacts across all of them step 5 -> enforce permissions which tools, how many writes, what needs approval. safety lives in the harness, never in the model step 6 -> wire observability track every tool call, cost and retry. trip wires fire when behavior drifts the result: your agent turns from a demo into infrastructure every failure makes the system permanently better, not just the next conversation that's also the line between an agent you show people and one a client pays you to leave running one harness gets you a reliable loop. connect several and it becomes graph engineering, where one agent's failure turns into another agent's sensor the PDF stops at the harness. the graph is the next layer that breakdown, running on Kimi, is below ↓

Annatar.md

78,571 views • 13 days ago

Anthropic hired this engineer at $250K-$750K a year because he knows how to build harnesses for multi-agent systems In this 15-minute workshop, he shows exactly how to build one from scratch AI → Agents → Harness → Loops → Graphs step 1 → start with the Claude Agent SDK - the harness handles loops, context, and sandboxing step 2 → separate the brain from the hands - reasoning in one place, tools in a sandbox, 60% faster to first token step 3 → run it server-side and log every step - close your laptop and it keeps running, crashes resume from the log step 4 → make failure cheap - retry dead sandboxes and replay lost context instead of starting over step 5 → turn yesterday's logs into new memory and skills - the harness wakes up smarter Anthropic calls this "dreaming" Most people spend weeks building this by hand You don't have to Bookmark and watch it Then read the full harness engineering guide below ↓

Anthropic hired this engineer at $250K-$750K a year because he knows how to build harnesses for multi-agent systems In this 15-minute workshop, he shows exactly how to build one from scratch AI → Agents → Harness → Loops → Graphs step 1 → start with the Claude Agent SDK - the harness handles loops, context, and sandboxing step 2 → separate the brain from the hands - reasoning in one place, tools in a sandbox, 60% faster to first token step 3 → run it server-side and log every step - close your laptop and it keeps running, crashes resume from the log step 4 → make failure cheap - retry dead sandboxes and replay lost context instead of starting over step 5 → turn yesterday's logs into new memory and skills - the harness wakes up smarter Anthropic calls this "dreaming" Most people spend weeks building this by hand You don't have to Bookmark and watch it Then read the full harness engineering guide below ↓

rari

183,635 views • 19 days ago