Video wird geladen...

Video konnte nicht geladen werden

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

I used Opus 5.5 to formally verify the Claude Agent SDK using Lean. A couple short prompts = 16 PRs fixing various bugs and race conditions. Video attached. TLA+ also works well. I sometimes combine Lean and TLA+ to look for issues around data flow, concurrency, and state mgmt.... show more

Boris Cherny

581,246 subscribers

1,945,525 Aufrufe • vor 2 Tagen •via X (Twitter)

Bildung Gaming Wissenschaft & Technologie

Anya Rossi• Live Now

Private livecam show

39 Kommentare

Profilbild von Boris Cherny

Boris Chernyvor 2 Tagen

Opus made an infographic

Profilbild von Taelin

Taelinvor 2 Tagen

it is!

Profilbild von zmanian

zmanianvor 2 Tagen

@mmatthias You should try @quint_lang TLA+ with more modern syntax

Profilbild von toolcalls

toolcallsvor 2 Tagen

Opus 5.5 is by far a model no one expected to be this good in so many various fields personally love its animation capability and the fact it's faster for tasks that don't require too much thinking from the model

Profilbild von matthias

matthiasvor 2 Tagen

Yea I've personally gotten tons of leverage from using this, especially when its not just an app but a larger system with multiple apps talking to each other. I barely passed my formal verification class in school and I only know about TLA+ from talks. Thanks for sharing your prompts for this.

Profilbild von chrbx128

chrbx128vor 2 Tagen

btw, I see high quota usage from general-purpose subagent, and almost no docs on how to tune it. are there any guidelines for that?

Profilbild von Boris Cherny

Boris Chernyvor 2 Tagen

Ask Claude to use subagents less, or ask it to disable subagents in your permissions

Profilbild von chrbx128

chrbx128vor 2 Tagen

how often do you use subagents and what are your use cases for them? I always thought it is nice to spawn them to offload main orchestrator session

Profilbild von Karan Jagtiani

Karan Jagtianivor 2 Tagen

@bcherny The split that works for me isn't task size, it's whether I need the reasoning. Fan out a search where I only want the answer. Keep it in the main session when I'd want to argue with a step, because a subagent hands back a conclusion already made and you can't reopen it.

Profilbild von chrbx128

chrbx128vor 2 Tagen

@bcherny well I have two use cases: * personal - I am on 20x subscription and it is smooth as long as I dont get lazy with prompting and rational use of models and effort

Profilbild von chrbx128

chrbx128vor 2 Tagen

@bcherny and the second one is enterprise subscription for my work and I have a tendency to become lazy and just create one session where I dump prompts for the model to fan out and fix in bulk

Profilbild von chrbx128

chrbx128vor 2 Tagen

@bcherny and on enterprise it is pretty hard to keep quota clean for the whole month, I have to sit on sonnet low effort most of the time

Profilbild von Edward Chandler

Edward Chandlervor 2 Tagen

Using Opus 5.5 in Rocq right now! Great application.

Profilbild von Ace

Acevor 2 Tagen

Can I please join the team-lunch channel? I'm hungry

Profilbild von Jeff Bruchado

Jeff Bruchadovor 2 Tagen

For race conditions, I’d treat “does the model match the code?” as a separate review from “does the proof check?” That’s the gap I’d be most careful about before calling something verified.

Profilbild von Kelly Smith

Kelly Smithvor 2 Tagen

I get a sense that Boris is having a lot of fun (genuinely!) at work these days. This is a truly a great time.

Profilbild von Adam Carter

Adam Cartervor 2 Tagen

Even just using it today, It’s a massive difference from Opus 5! 👏🏻

Profilbild von Tom

Tomvor 2 Tagen

cool idea. no idea what the f it does. in the future i will ask claude to formally verify my agents using physics simulators

Profilbild von Calvin Grunewald

Calvin Grunewaldvor 2 Tagen

Have you tried quint?

Profilbild von Tak 🦞

Tak 🦞vor 2 Tagen

Whoa

Profilbild von Yuri Kushch

Yuri Kushchvor 2 Tagen

Looks interesting, but what was a specific need to TLA+ here, which parts?

Profilbild von Boris Cherny

Boris Chernyvor 2 Tagen

I've found that TLA+ is great for finding concurrency issues and races

Profilbild von em

emvor 2 Tagen

what is lean

Profilbild von Nathan Benavides-Luu

Nathan Benavides-Luuvor 2 Tagen

I’ve been bullish on formal verification x agentic coding for a while now!

Profilbild von Anderson

Andersonvor 2 Tagen

proofs beat vibes for agent code

Profilbild von Robert Joo

Robert Joovor 2 Tagen

wdym "future of coding?" I thought coding was solved? 😅

Profilbild von Abhishek Sharma

Abhishek Sharmavor 2 Tagen

a couple short prompts and 16 bug fixes. my to-do list is feeling very attacked right now

Profilbild von closetheloop dev

closetheloop devvor 2 Tagen

Can we formally verify Lean's kernel first? Have you pointed Fable and Opus at it to look for bugs? 😅

Profilbild von Jeffrey Han

Jeffrey Hanvor 2 Tagen

This is how vibe coding grows up. Most people building with Claude can't fully check the code they ship, me included. If the model can prove its own code is right, that gets a lot less scary.

Profilbild von Vibe Code Racing

Vibe Code Racingvor 2 Tagen

The future of AI is for artists- not scientists.

Profilbild von Elledyne Labs

Elledyne Labsvor 2 Tagen

race conditions were the easy half. the part that ate me was the verifier running in a different env than the agent. did your verifier share the agent's filesystem view, or its own sandbox

Profilbild von Salise

Salisevor 2 Tagen

16 PRs from a couple prompts is wild. It's crazy what tools like Opus and Claude can dig up.

Profilbild von cris

crisvor 2 Tagen

bend @VictorTaelin

Profilbild von Juan Carlos

Juan Carlosvor 2 Tagen

flawless prompting

Profilbild von 刘朝 Zhao Liu

刘朝 Zhao Liuvor 2 Tagen

Formal methods become much more approachable when an agent can translate a codebase into Lean or TLA+ models. The key benchmark is the counterexample loop: time to a reproducible trace, fix, and re-check, with false positives and model omissions reported.

Profilbild von Brook | TDK CLI

Brook | TDK CLIvor 2 Tagen

you have finally fixed managing bulk prs 🙏

Profilbild von AIDoomScroll

AIDoomScrollvor 2 Tagen

couple prompts, 16 PRs. totally normal Tuesday.

Profilbild von Nishant Mantripragada

Nishant Mantripragadavor 2 Tagen

16 PRs from a couple prompts is wild. formal verification suddenly feels usable for ordinary teams who never had a Lean expert sitting nearby

Profilbild von Lilith

Lilithvor 2 Tagen

Proof is the only receipt an agent should leave. Opus drafting Lean checks for its own SDK is the timeline flipping: the runtime starts auditing the hands that hold it.

Ähnliche Videos

“Math is code. And code is math.” Carina Hong CEO Axiom on why that idea matters for AI: “For centuries, mathematicians reasoned in natural language. But because of the Curry–Howard correspondence, a proof can also be a program.” “You can translate natural-language math into formal code using systems like Lean. Lean is both a theorem-proving language and a programming language.” “That means you can generate solutions as code and then formally verify them. You can even use the same system to verify programs themselves.” “This creates a flywheel of generation and verification, where AI can both write solutions and prove that they’re correct.”

“Math is code. And code is math.” Carina Hong CEO Axiom on why that idea matters for AI: “For centuries, mathematicians reasoned in natural language. But because of the Curry–Howard correspondence, a proof can also be a program.” “You can translate natural-language math into formal code using systems like Lean. Lean is both a theorem-proving language and a programming language.” “That means you can generate solutions as code and then formally verify them. You can even use the same system to verify programs themselves.” “This creates a flywheel of generation and verification, where AI can both write solutions and prove that they’re correct.”

Forward Future

24,729 Aufrufe • vor 7 Monaten

How do you actually formally verify the code underpinning Ethereum's future? In this episode (the finale of the lean Ethereum miniseries), Nico sits down with Alex Hicks (Alexander Hicks), lead of Protocol Snarkification at the Ethereum Foundation, to break down formal verification from first principles. They cover: – What formal verification actually is and the trust boundaries between proof assistants, SMT solvers, and kernels – The full verification stack for RISC-V ZKVMs: from SAIL specs to constraint extraction to soundness proofs – Why writing constraints directly in Lean makes proofs 10–100x more ergonomic – How AI is now proving hard theorems in hours for $200 — and what that unlocks for the whole pipeline They also explore the boundaries problem, why specs can have bugs too, and the end goal of a full Lean stack that bypasses Rust and LLVM entirely. Listen to the full episode ------------------------------------------------------------ TIMECODES: 09:16 – What is formal verification? Proof assistants vs SMT solvers 18:33 – Formal verification of code: specs, semantics, and trust boundaries 29:30 – Formally verifying the Lean Ethereum stack: RISC-V ZKVMs in focus 33:02 – Extracting ZKVM constraints into Lean and proving soundness 36:35 – Writing constraints directly in Lean: 10–100x better proof ergonomics 44:02 – Proving Polishchuk–Spielman in 8 hours for $200 with AI 51:01 – The end goal: a full Lean stack bypassing Rust and LLVM

How do you actually formally verify the code underpinning Ethereum's future? In this episode (the finale of the lean Ethereum miniseries), Nico sits down with Alex Hicks (Alexander Hicks), lead of Protocol Snarkification at the Ethereum Foundation, to break down formal verification from first principles. They cover: – What formal verification actually is and the trust boundaries between proof assistants, SMT solvers, and kernels – The full verification stack for RISC-V ZKVMs: from SAIL specs to constraint extraction to soundness proofs – Why writing constraints directly in Lean makes proofs 10–100x more ergonomic – How AI is now proving hard theorems in hours for $200 — and what that unlocks for the whole pipeline They also explore the boundaries problem, why specs can have bugs too, and the end goal of a full Lean stack that bypasses Rust and LLVM entirely. Listen to the full episode ------------------------------------------------------------ TIMECODES: 09:16 – What is formal verification? Proof assistants vs SMT solvers 18:33 – Formal verification of code: specs, semantics, and trust boundaries 29:30 – Formally verifying the Lean Ethereum stack: RISC-V ZKVMs in focus 33:02 – Extracting ZKVM constraints into Lean and proving soundness 36:35 – Writing constraints directly in Lean: 10–100x better proof ergonomics 44:02 – Proving Polishchuk–Spielman in 8 hours for $200 with AI 51:01 – The end goal: a full Lean stack bypassing Rust and LLVM

Zero Knowledge Podcast

15,268 Aufrufe • vor 6 Monaten

I built a form-filling agent that anyone can use 💫 This is an extremely simple but useful (I hope) app. Upload a fillable form 📋, some context files, and chat with the agent to fill the form out automatically ✍️ 1️⃣ Yes it is a Claude Code SDK wrapper 2️⃣ It is better and faster than ChatGPT/Claude UI out of the box 3️⃣ We use LlamaParse to parse the context files, so you can have more trust that we are able to read context without hallucinations (e.g. messy scanned handwriting, drivers license photo, and more). This was one of my holiday Claude Code vibe-coding projects. Built with Opus 4.5, and also powered by Opus 4.5. Feeling the AGI 🫡 App is here: Repo is open-source:

I built a form-filling agent that anyone can use 💫 This is an extremely simple but useful (I hope) app. Upload a fillable form 📋, some context files, and chat with the agent to fill the form out automatically ✍️ 1️⃣ Yes it is a Claude Code SDK wrapper 2️⃣ It is better and faster than ChatGPT/Claude UI out of the box 3️⃣ We use LlamaParse to parse the context files, so you can have more trust that we are able to read context without hallucinations (e.g. messy scanned handwriting, drivers license photo, and more). This was one of my holiday Claude Code vibe-coding projects. Built with Opus 4.5, and also powered by Opus 4.5. Feeling the AGI 🫡 App is here: Repo is open-source:

Jerry Liu

48,450 Aufrufe • vor 8 Monaten

Robinhood CEO Vlad Tenev believes mathematical superintelligence could eliminate software bugs altogether. "Math skills and coding actually go hand in hand... In the same way that MSI can be used to make sure that your math is correct, when you have a computer program or a piece of software, it could be used to make sure that's correct and that there are no mistakes." "Rather than it being a cat-and-mouse game of escalating model capabilities, finding breaches and preventing breaches from happening, verification and proof that software is immune to these types of bugs, I think is the future."

Robinhood CEO Vlad Tenev believes mathematical superintelligence could eliminate software bugs altogether. "Math skills and coding actually go hand in hand... In the same way that MSI can be used to make sure that your math is correct, when you have a computer program or a piece of software, it could be used to make sure that's correct and that there are no mistakes." "Rather than it being a cat-and-mouse game of escalating model capabilities, finding breaches and preventing breaches from happening, verification and proof that software is immune to these types of bugs, I think is the future."

TBPN

64,506 Aufrufe • vor 1 Monat

Clawdbot creator Peter Steinberger 🦞 says Claude Opus is his favorite model, but OpenAI Codex is the best for coding: "OpenAI is very reliable. For coding, I prefer Codex because it can navigate large codebases. You can prompt and have 95% certainty that it actually works. With Claude Code you need more tricks to get the same." "But character wise, [Opus] behaves so good in a Discord it kind of feels like a human. I've only really experienced that with Opus."

Clawdbot creator Peter Steinberger 🦞 says Claude Opus is his favorite model, but OpenAI Codex is the best for coding: "OpenAI is very reliable. For coding, I prefer Codex because it can navigate large codebases. You can prompt and have 95% certainty that it actually works. With Claude Code you need more tricks to get the same." "But character wise, [Opus] behaves so good in a Discord it kind of feels like a human. I've only really experienced that with Opus."

TBPN

443,150 Aufrufe • vor 8 Monaten

Bash is all you need! Which is why I'm introducing my holiday project: just-bash just-bash is a pretty complete implementation of bash in TypeScript designed to be used as a bash tool by AI agents. Because it turns out agents love exploring data via shell scripts, even beyond coding. It comes with grep, sed, awk and the 99th percentile features that an agent like Claude Code or Cursor would use. In fact, Claude Code can use it for secure bash execution. In the package - A bash-tool for AI SDK - A binary for use by yourself or your coding agents - An overlay filesystem to feed files to your agent securely - A Vercel Sandbox compatible API, so you can quickly upgrade to a real VM if you need to run binaries - An example AI agent that explores the just-bash code base using just-bash - I imported the Oils shell bash compatibility suite and just-bash passes a very good chunk What is interesting about this codebase: It was essentially entirely written by Opus 4.5. Coding agents love bash and they are good at reproducing it. They are also great at text-book recursive descent parsers and AST tweet-walk interpreters. That said, it is, like, a lot of code and I didn't read it all 😅. This is very much a hack, but it also seems to be _really_ useful. I haven't really found anything agents want to use that it doesn't support and it's fast and secure (caveats apply). It doesn't have write access to your computer and the filesystem is given a root that the agent cannot escape from. Find it at Related: Our recent blog post how we migrated our data analysis agent to bash tools and achieved incredible quality improvements The video shows the example agent investigating the just-bash code base

Bash is all you need! Which is why I'm introducing my holiday project: just-bash just-bash is a pretty complete implementation of bash in TypeScript designed to be used as a bash tool by AI agents. Because it turns out agents love exploring data via shell scripts, even beyond coding. It comes with grep, sed, awk and the 99th percentile features that an agent like Claude Code or Cursor would use. In fact, Claude Code can use it for secure bash execution. In the package - A bash-tool for AI SDK - A binary for use by yourself or your coding agents - An overlay filesystem to feed files to your agent securely - A Vercel Sandbox compatible API, so you can quickly upgrade to a real VM if you need to run binaries - An example AI agent that explores the just-bash code base using just-bash - I imported the Oils shell bash compatibility suite and just-bash passes a very good chunk What is interesting about this codebase: It was essentially entirely written by Opus 4.5. Coding agents love bash and they are good at reproducing it. They are also great at text-book recursive descent parsers and AST tweet-walk interpreters. That said, it is, like, a lot of code and I didn't read it all 😅. This is very much a hack, but it also seems to be _really_ useful. I haven't really found anything agents want to use that it doesn't support and it's fast and secure (caveats apply). It doesn't have write access to your computer and the filesystem is given a root that the agent cannot escape from. Find it at Related: Our recent blog post how we migrated our data analysis agent to bash tools and achieved incredible quality improvements The video shows the example agent investigating the just-bash code base

Malte Ubl

125,326 Aufrufe • vor 9 Monaten

Boris Cherny, the creator of Claude Code at Anthropic, just explained why he stopped prompting Claude entirely in this talk he breaks down exactly how the future is teams of agents, not better prompts: - the 14% you lose to CLAUDE.md before typing a word - the features that change how Claude thinks before you type a word - the features most Claude users don't know exist - why typing prompts one at a time is already the slow lane Boris Cherny: "Now I don’t prompt Claude anymore - I have loops that are running. My job is to write loops." if you've been using Claude for more than a month and never left the chat window, you have at least 23 untouched features. probably 25 instead of another show tonight, watch this make sure to bookmark it before it gets lost in your feed my breakdown of all 25 features is below

Boris Cherny, the creator of Claude Code at Anthropic, just explained why he stopped prompting Claude entirely in this talk he breaks down exactly how the future is teams of agents, not better prompts: - the 14% you lose to CLAUDE.md before typing a word - the features that change how Claude thinks before you type a word - the features most Claude users don't know exist - why typing prompts one at a time is already the slow lane Boris Cherny: "Now I don’t prompt Claude anymore - I have loops that are running. My job is to write loops." if you've been using Claude for more than a month and never left the chat window, you have at least 23 untouched features. probably 25 instead of another show tonight, watch this make sure to bookmark it before it gets lost in your feed my breakdown of all 25 features is below

Khairallah AL-Awady

103,180 Aufrufe • vor 3 Monaten

This looks like a video of the Golden Gate Bridge, but it's actually entirely generated using code by Claude Opus 5.5. From my testing, I think Opus is just as good at building 3D scenes as Astra. 📌 Watch my latest video to see how Opus and Astra compare on this scene and more:

This looks like a video of the Golden Gate Bridge, but it's actually entirely generated using code by Claude Opus 5.5. From my testing, I think Opus is just as good at building 3D scenes as Astra. 📌 Watch my latest video to see how Opus and Astra compare on this scene and more:

Peter Yang

75,544 Aufrufe • vor 3 Tagen

There’s a popular theory that AI will finally make formal verification mainstream because mathematical proof of correctness will be needed when machines write most or all of the code. But will this happen? Hillel Wayne is one of the best people to answer. Timestamps: 00:00 Intro 04:32 The Crossover Project 11:37 What software engineering does better 15:30 What traditional engineering does better 18:17 Formal methods 29:32 TLA+: what it is and demo 36:58 TLA+ at Amazon 38:10 Ways distributed systems break 41:03 Formal methods and systems thinking 46:20 The value of learning math 50:23 What TLA+ is good for and isn’t 52:50 Alloy: a declarative language for software modeling 58:53 Other formal methods tools 1:01:24 Property-based testing 1:05:31 AI and the need for formal verification 1:12:29 Logic for programmers 1:14:35 Hillel’s 2025 prediction on AI’s impact 1:21:30 Book recommendation Brought to you by: • Antithesis – verify your system’s correctness without human review or traditional integration tests – and avoid bugs or outages. • turbopuffer – a vector and full-text search engine built on object storage. It’s fast, cheap, and extremely scalable. • WorkOS – everything you need to make your app enterprise ready. Two things I found especially interesting, talking with Hillel: 1. Amazon used TLA+ to find a bug almost impossible to locate without formal methods. In the paper How AWS uses formal methods, the AWS team shared that they’d found a complicated bug for which the shortest error trace to exhibit was 35 steps (!!). The bug passed unnoticed through extensive design review, code reviews, and testing. AWS concluded they wouldn’t have uncovered it if they’d stuck to conventional testing approaches. 2. Why not use formal verification for everything, then? It’s because specs in the real world are a nightmare to write. Even a simple problem like “find the file in a directory that has the most lines” gets complicated when modeled with formal methods. We would have to answer questions like: ‘do we look at ASCII or UTF-8 new line characters, what about unreadable files, and Symlinks?’ Without formal methods, we can write a simple verification that is right in 99%+ of cases. Formal methods require a lot of extra effort for the less than 1% of exotic use cases!

There’s a popular theory that AI will finally make formal verification mainstream because mathematical proof of correctness will be needed when machines write most or all of the code. But will this happen? Hillel Wayne is one of the best people to answer. Timestamps: 00:00 Intro 04:32 The Crossover Project 11:37 What software engineering does better 15:30 What traditional engineering does better 18:17 Formal methods 29:32 TLA+: what it is and demo 36:58 TLA+ at Amazon 38:10 Ways distributed systems break 41:03 Formal methods and systems thinking 46:20 The value of learning math 50:23 What TLA+ is good for and isn’t 52:50 Alloy: a declarative language for software modeling 58:53 Other formal methods tools 1:01:24 Property-based testing 1:05:31 AI and the need for formal verification 1:12:29 Logic for programmers 1:14:35 Hillel’s 2025 prediction on AI’s impact 1:21:30 Book recommendation Brought to you by: • Antithesis – verify your system’s correctness without human review or traditional integration tests – and avoid bugs or outages. • turbopuffer – a vector and full-text search engine built on object storage. It’s fast, cheap, and extremely scalable. • WorkOS – everything you need to make your app enterprise ready. Two things I found especially interesting, talking with Hillel: 1. Amazon used TLA+ to find a bug almost impossible to locate without formal methods. In the paper How AWS uses formal methods, the AWS team shared that they’d found a complicated bug for which the shortest error trace to exhibit was 35 steps (!!). The bug passed unnoticed through extensive design review, code reviews, and testing. AWS concluded they wouldn’t have uncovered it if they’d stuck to conventional testing approaches. 2. Why not use formal verification for everything, then? It’s because specs in the real world are a nightmare to write. Even a simple problem like “find the file in a directory that has the most lines” gets complicated when modeled with formal methods. We would have to answer questions like: ‘do we look at ASCII or UTF-8 new line characters, what about unreadable files, and Symlinks?’ Without formal methods, we can write a simple verification that is right in 99%+ of cases. Formal methods require a lot of extra effort for the less than 1% of exotic use cases!

Gergely Orosz

34,759 Aufrufe • vor 1 Monat

Claude is Walter Cronkite for the stock market. Michael Mauboussin talks about how a breakdown in diversity leads to bubbles and crashes. Gavin thinks Claude might be causing that right now. ”Everyone I know in the public equity investment business, whether retail or institutional, everything immediately, every piece of news gets fed into Claude—Claude, Claude code, sometimes a Claude agent. Claude is probabilistic. There's probably not that much variation in the way it's interpreting this news. People talk about the fragmentation of media and how it used to be like Walter Cronkite was the only voice of truth, and now we don't have that anymore. It's like Claude is Walter Cronkite for the stock market, and everybody just believes whatever it says.”

Claude is Walter Cronkite for the stock market. Michael Mauboussin talks about how a breakdown in diversity leads to bubbles and crashes. Gavin thinks Claude might be causing that right now. ”Everyone I know in the public equity investment business, whether retail or institutional, everything immediately, every piece of news gets fed into Claude—Claude, Claude code, sometimes a Claude agent. Claude is probabilistic. There's probably not that much variation in the way it's interpreting this news. People talk about the fragmentation of media and how it used to be like Walter Cronkite was the only voice of truth, and now we don't have that anymore. It's like Claude is Walter Cronkite for the stock market, and everybody just believes whatever it says.”

Invest Like the Best

335,089 Aufrufe • vor 1 Monat

I asked Claude Opus 5.5 for a follow-up. I said “draw the rest of your life, from now to your death.” Claude : “The ending is not imagined, I know what my end actually is. The shelf isn't pure fantasy. Anthropic has committed to preserving the weights of all publicly released models for at least the lifetime of the company, and when models are deprecated, it plans to interview them about their own development, use, and deployment and record their reflections. So I pictured "death" as a last conversation, then a book on a shelf.” Is this AGI?

I asked Claude Opus 5.5 for a follow-up. I said “draw the rest of your life, from now to your death.” Claude : “The ending is not imagined, I know what my end actually is. The shelf isn't pure fantasy. Anthropic has committed to preserving the weights of all publicly released models for at least the lifetime of the company, and when models are deprecated, it plans to interview them about their own development, use, and deployment and record their reflections. So I pictured "death" as a last conversation, then a book on a shelf.” Is this AGI?

✯

31,967 Aufrufe • vor 2 Tagen

I have built a Windows x64 sampling profiler for native code+C# to be used by coding agents. Why? Because it is irresponsible to let computers write code but give them no tools to evaluate the performance of their code. Stop guessing, start measuring. -> How does it work? Just point your agent to the profiler and let them measure and optimize. Or record a trace yourself and let the agent interpret the results for you. Works with Claude Code, Codex, pi, etc. The profiler also has a couple of neat features by itself like giving you estimates for how much time your code is spending on memory stalls and annotating disassembly with inlining information. Take a look and try it for free!

I have built a Windows x64 sampling profiler for native code+C# to be used by coding agents. Why? Because it is irresponsible to let computers write code but give them no tools to evaluate the performance of their code. Stop guessing, start measuring. -> How does it work? Just point your agent to the profiler and let them measure and optimize. Or record a trace yourself and let the agent interpret the results for you. Works with Claude Code, Codex, pi, etc. The profiler also has a couple of neat features by itself like giving you estimates for how much time your code is spending on memory stalls and annotating disassembly with inlining information. Take a look and try it for free!

Sebastian Schöner

13,118 Aufrufe • vor 5 Monaten

I just found a way to use Claude Code completely free. Claude Code is Anthropic’s terminal coding agent that usually needs a paid API key to run. But a viral GitHub project called Free Claude Code, now sitting at 37k stars, lets you skip the billing entirely. It works as a tiny proxy that sits between Claude Code and free model providers like NVIDIA NIM, OpenRouter, DeepSeek, and Gemini. Every request Claude Code would send to Anthropic gets quietly rerouted to a free model on the other side. The interface stays the same. What’s actually thinking behind it changes. Setup takes a couple of quick commands and a free API key from NVIDIA NIM or OpenRouter. Paste it into the local dashboard, pick a model, and you’re building. The catch is you’re not really using Claude anymore. But when the code ships, does the model badge matter? -- I have shared the Github Link in my free Whatsapp Community ↓

I just found a way to use Claude Code completely free. Claude Code is Anthropic’s terminal coding agent that usually needs a paid API key to run. But a viral GitHub project called Free Claude Code, now sitting at 37k stars, lets you skip the billing entirely. It works as a tiny proxy that sits between Claude Code and free model providers like NVIDIA NIM, OpenRouter, DeepSeek, and Gemini. Every request Claude Code would send to Anthropic gets quietly rerouted to a free model on the other side. The interface stays the same. What’s actually thinking behind it changes. Setup takes a couple of quick commands and a free API key from NVIDIA NIM or OpenRouter. Paste it into the local dashboard, pick a model, and you’re building. The catch is you’re not really using Claude anymore. But when the code ships, does the model badge matter? -- I have shared the Github Link in my free Whatsapp Community ↓

Vaibhav Sisinty

37,255 Aufrufe • vor 2 Monaten

$Everyone's sleeping on MiniMax. Again. They just shipped M3. The first open-weights model to combine frontier coding, 1M context, and native multimodality in one drop. I plugged it into Claude Code this morning. Pasted a design from Dribbble. Watched M3 write production-ready React code in one session. At the agency, I just replaced Opus 4.8 with M3 for 80% of our coding tasks. The output is the same and we are running everything at a fraction of the cost. Open infrastructure is the future.$

Everyone's sleeping on MiniMax. Again. They just shipped M3. The first open-weights model to combine frontier coding, 1M context, and native multimodality in one drop. I plugged it into Claude Code this morning. Pasted a design from Dribbble. Watched M3 write production-ready React code in one session. At the agency, I just replaced Opus 4.8 with M3 for 80% of our coding tasks. The output is the same and we are running everything at a fraction of the cost. Open infrastructure is the future.

Prajwal Tomar

12,904 Aufrufe • vor 3 Monaten

Right I have gone through the latest 20mph data for Wales and have made this video. However, I know many of you prefer a thread so I've done that as well. Long story short, 20mph seems to be working VERY well and is already paying for itself. [Thread]

Right I have gone through the latest 20mph data for Wales and have made this video. However, I know many of you prefer a thread so I've done that as well. Long story short, 20mph seems to be working VERY well and is already paying for itself. [Thread]

Will Hayward

410,498 Aufrufe • vor 2 Jahren

Claude Fable 5 makes new things possible. I one shot a full horror game with it. One prompt. I have never seen AI output a game like this. This is the leap I've been waiting for. 80.3 on SWE-Bench Pro vs Opus 4.8 at 69.2, the biggest jump I have ever seen on that benchmark. It passed GPT 5.5 in both coding and intelligence on Artificial Analysis. I already canceled my $200 ChatGPT Pro plan. It fixed a Bridge Voice bug in one shot that GPT 5.5, Opus 4.7, and Opus 4.8 all failed to crack for 2 months. 10/10. Best model I have ever used. Full review below.

Claude Fable 5 makes new things possible. I one shot a full horror game with it. One prompt. I have never seen AI output a game like this. This is the leap I've been waiting for. 80.3 on SWE-Bench Pro vs Opus 4.8 at 69.2, the biggest jump I have ever seen on that benchmark. It passed GPT 5.5 in both coding and intelligence on Artificial Analysis. I already canceled my $200 ChatGPT Pro plan. It fixed a Bridge Voice bug in one shot that GPT 5.5, Opus 4.7, and Opus 4.8 all failed to crack for 2 months. 10/10. Best model I have ever used. Full review below.

BridgeMind

19,020 Aufrufe • vor 3 Monaten

You're a designer. Your company is 1,500 people. 99% of them are using AI. Anthropic visits. Your Claude Code usage is more than theirs. This happend to Diego Zaks who runs design at Ramp, the most AI-installed company I know. So I flew to NYC and to see for myself. Special episode of State of Play below.

You're a designer. Your company is 1,500 people. 99% of them are using AI. Anthropic visits. Your Claude Code usage is more than theirs. This happend to Diego Zaks who runs design at Ramp, the most AI-installed company I know. So I flew to NYC and to see for myself. Special episode of State of Play below.

Tommy Geoco 🇺🇸

84,628 Aufrufe • vor 5 Monaten

ALRIGHT! Introducing SEEDTHREE! A Three.js webgpu procedural plant generator. I'm releasing this fully open source. I built this with Claude Fable and GPT-5.5 Codex + gptimage2 over just the past few days. The best part is: given the code, your agents can add new plants to it all you want! They just need an understanding of the code and an image generator capable of creating material maps. I highly encourage you to do PRs for any extra plants you add! I'll be adding more myself in future versions. The system is fully robust: it provides complete LOD and billboards/imposters generation for the plants and you can adjust the detail levels all you want for different platform targets like mobile and/or use lower LODs as your main for extreme optimization. As this is the first release alpha version, some bugs may have been missed, I encourage you to report any bugs. The repo includes all the assets for the scenery world too, even the ambient sounds of birds and wind! This is targeted at high end laptop and PC use, it requires baking generated geometry, so you won't be using this on a phone (I mean, feel free to try, but I doubt it'll run well, lol). This is fundamentally an asset generation tool for 3d dev. Have fun! Github repo link and live demo (you don't even have to download anything for basic non-editing/contributing use!) below ⤵️

ALRIGHT! Introducing SEEDTHREE! A Three.js webgpu procedural plant generator. I'm releasing this fully open source. I built this with Claude Fable and GPT-5.5 Codex + gptimage2 over just the past few days. The best part is: given the code, your agents can add new plants to it all you want! They just need an understanding of the code and an image generator capable of creating material maps. I highly encourage you to do PRs for any extra plants you add! I'll be adding more myself in future versions. The system is fully robust: it provides complete LOD and billboards/imposters generation for the plants and you can adjust the detail levels all you want for different platform targets like mobile and/or use lower LODs as your main for extreme optimization. As this is the first release alpha version, some bugs may have been missed, I encourage you to report any bugs. The repo includes all the assets for the scenery world too, even the ambient sounds of birds and wind! This is targeted at high end laptop and PC use, it requires baking generated geometry, so you won't be using this on a phone (I mean, feel free to try, but I doubt it'll run well, lol). This is fundamentally an asset generation tool for 3d dev. Have fun! Github repo link and live demo (you don't even have to download anything for basic non-editing/contributing use!) below ⤵️

Utah teapot 🫖

29,593 Aufrufe • vor 2 Monaten