~BBA

software engineer, investor, addicted to information

[Andrew Ng @AndrewYNg]

“Loop engineering” is a hot buzzphrase after mentions of it by Boris Cherny (Claude Code’s creator) and Peter Steinberger (OpenClaw's creator) went viral on social media. Loops are now a key part of how we get AI agents to iterate at length to build software. In this letter, I’d like to share my 3 key loops, shown in the image below, for building 0-to-1 products. These loops guide not just how I build software, but also how I decide what software to build.

Agentic coding loop: Given a product specification and optionally a set of evals (that is, a dataset against which to measure performance), we can have an AI agent write code, test its work, and keep iterating until the code is bug-free and meets its specification. This idea of closing the loop took off around the end of last year, and it has been a game changer in enabling coding agents to work longer productively without human intervention. For example, over the weekend, I was building an app for my daughter to practice typing, and my coding agent could easily work for around an hour, using a web browser to check what it had built multiple times before getting back to me, without needing my intervention.

The engineering loop executes quickly. Every few minutes, the coding agent might build and test a new version of the software. I hear frequently from developers who are finding new ways to engineer more effective engineering loops. This is an active area of invention!

Developer feedback loop: In this loop, a developer examines the current product and steers the coding agent to improve it. Last year, a lot of developers (including me) were acting as the QA (quality assurance) function for our coding agents, manually finding bugs and then asking the agent to fix them. But with coding agents much more able to test their own code, the amount of time we need to spend on this function has decreased significantly. This allows us to make higher-level product decisions, such as what key features to offer, where the UI needs improvement, and so on.

The developer-feedback loop operates over time intervals between tens of minutes and hours — that's how frequently a developer might review a product and give feedback. In the case of the typing app, I changed my mind a few times about the visual design, what cat costumes she can unlock as she learns (she loves cats), and the user flow for a grown-up to log in and steer the child's learning experience.

When a developer has a clear vision for what to build, it is still a lot of work to translate that vision into a specification for a coding agent to implement. Further, after the developer has seen an implementation, they might update (or perhaps clarify) the spec to steer it toward what they want. If you find that the system repeatedly runs into certain problems, building a set of evals for the agent becomes useful.

AI-native teams are increasingly using AI to help shape product direction, for example, automating the gathering and analysis of usage data, summarizing written and verbal customer feedback, or carrying out competitive analysis. However, for pretty much all the products I’m involved in, I see humans as having a significant context advantage over current AI systems — we know a lot more than the AI system about the users and the context the product has to operate in — and thus humans play a critical role. Many people describe this human contribution as “taste,” but I prefer to think of it as humans having a context advantage, since that gives us a clearer path to helping AI systems get better. This also speaks to why this step can’t be automated: So long as the human knows something the AI does not, human-in-the-loop is needed to to inject that knowledge into the system.

External feedback loop: This includes a wide range of tactics like asking a few friends for feedback, launching to alpha testers, or putting the code into production with A/B testing. These tactics are usually slow, rarely taking less than hours and sometimes taking days or even weeks. This data informs the developer vision, which in turn continues to drive the detailed product spec, which in turn drives the coding agent.

With coding agents speeding up software development, more engineers are starting to play a partial product management role. For many engineers who are growing into this role, the hardest part is shaping the product vision and striking a balance between building (bridging the gap between vision and spec) and getting user feedback to evolve the vision. It is important to do both!

I will write more about how to do this in future posts, but for now, I find it encouraging that engineers are playing an expanded role (just as product managers and designers now do more engineering).

[Original text: The Batch]

[Boris Cherny @bcherny]

As engineering, product, design, DS, etc. melt into a new kind of role, I was reflecting on what roles might look like in the future. For example, when I look at the Claude Code team I see what I think is five archetypes:

1. Prototyper: comes up with brand new ideas; churns out many ideas, most of which don't ship
2. Builder: quickly turns a prototype/idea into production-grade product/infra
3. Sweeper: cleans up the UI, simplifies the code and system, unships, optimizes performance
4. Grower: takes a product that has been built and iterates on it to improve Product-Market Fit
5. Maintainer: owns a mature system to make it secure, reliable, fast, and efficient as it scales

Many people span across 2 roles, and sometimes 3 roles. I also notice that these roles are not really tied to job function -- eg. across Anthropic, some designers match category 1, some 2, some 3; same for engineers, PM, DS.

A healthy team needs a mix of these, depending on the product:

- A product that is new and pre-PMF needs people that are strong at 1+2+3
- A product that is growing and has found PMF needs 2+3+4 and some 5
- A product that has strong PMF needs 3+4+5 and some 2

Maybe product roles of the future will look more like this, and less like the domain-specific roles of today?

[Gaurav @gauravsbuilding]

This is seriously terrifying..

[GREG ISENBERG @gregisenberg]

I just got back from SF and I FEEL INSPIRED.

I spent 5 days with frontier AI model teams, AI startup founders, and 3 billionaires.

My takeaways:

1. I had lunch with 3 billionaires. All of them are buying SaaS companies and rebuilding them agent-first. They were deeply inspired by Bending Spoons and Ryan Cohen's eBay deal. Buy the company, cut the headcount, rebuild the tech, add agents, add features, make more valuable experience, raise prices.

2. The frontier model companies are hungry for usage data from the field. They can see API calls and token counts. They can't see the actual workflows. If you're deep in a niche using these models in ways the model companies haven't seen, that understanding is incredibly valuable. Usage intelligence is the new alpha.

3. Consumer AI is massively underbuilt. Every billboard in SF is either B2B inference infrastructure or vertical agent companies. The entire city is optimized for enterprise. Meanwhile you have companies like Cal AI doing $50M ARR in 18 months as a consumer app. I met with a cool few teams doing consumer AI (@paulscherer / @ekuyda)

4. MCP came up in literally every conversation. The companies exposing their product as MCP endpoints are getting pulled into deals they never pitched for. The ones that aren't are becoming invisible to agents. This is the new SEO. If agents can't find you, you don't exist. Building products for agents is the new zeitgeist in general.

5. Not uncommon for hot seed rounds to be $25-50 million valuations. I saw a Series A at $450 million

6. If I had a dollar every time someone mentioned "forward-deployed engineer" this trip I could have funded a seed round. It's the hottest role in SF right now. The person who sits between the agent and the customer, making sure everything actually works.

7. The mood around open source shifted. A year ago it felt like open source was chasing the frontier models. Now founders are telling me Gemma and DeepSeek are good enough for 80% of what they need at a fraction of the cost. The "which model do you use" conversation is being replaced by "which model for which task." Model loyalty kinda feels dead.

8. Voice agents came up more than I expected. Multiple founders told me voice is the interface for the next billion users. The billion people who will never type a prompt will absolutely talk to one.

9. The Obsidian community in SF is weirdly intense. Multiple founders showed me their vaults unprompted. Like showing someone your home gym. It's a flex now. The quality of your knowledge base (second brain?) is becoming a status symbol among builders.

10. Maybe it was just the people I met but the age of the founders is shifting. I met more founders over 40 this trip than any trip before and more founders under age 21 than ever before. Founders getting older and younger at the same time.

11. I spoke to a lot of fast-growing startups, VCs and frontier models who are hiring content creators right now.

12. The restaurant scene in SF is actually better than it's been in years. Founders are going out more. Alcohol is out, not surprisingly.

13. SF doesn't feel like the only place anymore. We all have access to the same frontier models. We all read the same X feed. A founder in NYC or Lagos is calling the same APIs as a founder in SoMa. So in the past it felt like SF was always lightyears ahead, doesn't feel that way anymore. It's okay not to live in SF and have BIG DREAMS.

14. The coworking spaces in SF are half empty but the coffee shops are packed. People want to be around people. I had a few startup ideas here....

15. Walking around the Mission I noticed something: the street-level businesses, the taquerias, the barbershops, the laundromats, none of them use any AI at all.

16. I heard the phrase "agent debt" for the first time. Like technical debt but for agents. When you hack together an agent workflow fast and never clean it up, the system prompts conflict, the memory gets polluted, the tools overlap. 6 months later the agent is doing weird things and nobody knows why lol.

17. Met a few people who carry two phones now. One for personal. One that's basically an agent terminal running Telegram or iMessage connections to their agent fleet.

It's always amazing to get that dose of inspiration in SF. I FEEL INSPIRED.

But I'm so happy to be back home, locked in and building.

We're 12-18 months into a shift that will take 15 years to play out. The urgency in every conversation was real.

What an incredible time to be building.

[Elon Musk @elonmusk]

Regarding the OpenAI case, the judge & jury never actually ruled on the merits of the case, just on a calendar technicality.

There is no question to anyone following the case in detail that Altman & Brockman did in fact enrich themselves by stealing a charity. The only question is WHEN they did it!

I will be filing an appeal with the Ninth Circuit, because creating a precedent to loot charities is incredibly destructive to charitable giving in America.

OpenAI was founded to benefit all of humanity.

[Elon Musk @elonmusk]

Where will AI be in 1, 2 or 3 years?

[Rahul @sairahul1]

Karpathy didn't make a course.

He made THE course.

3 hours. Free.

Tokenization. Attention. Hallucinations. Tool use. RLHF. DeepSeek. AlphaGo.

Every behavior you've ever wondered about in an LLM - where it comes from, why it exists, how it was engineered.

The gap between engineers who understand this and engineers who don't isn't technical depth.

It's the ability to conceive of entirely different things.

[Reads with Ravi @readswithravi]

I’m in love with this sentence:

“Discipline looks boring until you see what it builds.”

[Chamath Palihapitiya @chamath]

This isn’t just a $META issue. It’s increasingly an issue for all of Tech. Because even within Tech, there is an emerging divide between AI-superpowered engineer/PM/sales folks and everyone else.

And we know how that will end: Smaller orgs, bigger payoffs but then the riches distributed across even fewer players that it is today. And everyone sees it coming. Hence the rot from the outside and, now, the inside.

To my peers and to the super entrepreneurs who are far more successful than me:

This is why elites in tech that “have made it” are increasingly the ones that EVERYONE hates. We’ve not distinguished the difference between luck and skill that got us here. We don’t act as stewards in the broadest sense of the term. We aren’t bringing society along like other generations of super successful business people have. We don’t pay it forward in any meaningful way - although we have clever ways to make it look like we do. Mostly, people see us hording all the gains.

Modern technology companies have essentially created a new form of indentured servitude for the educated masses where SBC was used to pay you an incredible living wage so you don’t go work “for the other guy”. But what does it mean to make $500k/yr if you still leak 55% to taxes then your landlord takes the next 25% for rent?? You still can’t buy a house in SV. These folks are on an eternal hamster wheel. It turns out it’s not much different than being in middle America making $55k.

[Nav Toor @heynavtoor]

🚨 OpenAI charges $0.006/minute. Google charges $0.024. AWS charges $0.024.

Someone just open sourced a tool that does it for $0. And it's faster than all of them.

It's called Insanely Fast Whisper. And that's not hype. That's the benchmark.

150 minutes of audio. 98 seconds to transcribe. On your own machine. No API key. No cloud. No per-minute billing.

Here's what the numbers look like:

→ Whisper Large v3 + Flash Attention 2: 150 min of audio in 98 seconds
→ Distil Whisper + Flash Attention 2: 150 min in 78 seconds
→ Standard Whisper without optimization: 31 minutes for the same job
→ That's a 19x speedup. Same model. Same accuracy. Just faster.

Here's what it does:

→ One command to transcribe any audio file or URL
→ Speaker diarization — knows WHO said WHAT
→ Transcription AND translation to other languages
→ Runs on NVIDIA GPUs and Mac (Apple Silicon)
→ Flash Attention 2 for maximum speed
→ Clean JSON output with timestamps
→ Works with every Whisper model variant

Here's the wildest part:

https://t.co/WfJGCpSz09 charges $100/year. Rev charges $1.50/minute. Descript charges $24/month. Enterprise transcription contracts cost thousands.

Podcasters, journalists, researchers, lawyers, content creators — anyone still paying for transcription is lighting money on fire.

8.8K GitHub stars. 633 forks. MIT License.

100% Open Source.

(Link in the comments)

[Tuki @TukiFromKL]

🚨 do you understand what Karpathy just said..

the guy who co-founded OpenAI.. led AI at Tesla.. one of the best engineers alive..

built an app with AI.. and said the code was the easy part..

the hard part was Stripe.. auth.. DNS.. databases.. deploying it.. connecting 15 different services that all have different dashboards and different docs and different billing pages..

AI can write your entire app in 20 minutes.. but it still can't click "confirm email" on Vercel..

so the thing that's "replacing developers" can't do the thing developers actually spend 80% of their time doing..

vibe coding didn't kill software engineering.. it just proved that coding was never the job.. the job was dealing with the mess around the code.. and that mess is still 100% human.

[Tuki @TukiFromKL]

🚨Did you see what Karpathy just said? Stop everything and read this.

3 years ago this man was teaching the world how to build neural networks from scratch.. Stanford lectures. YouTube tutorials.. "Here's how to write a GPT from zero."

> Now he's saying don't write code at all. Just manage the agents that write it for you.

The same guy who literally taught a generation of engineers how to code is now telling them their job is to sit in a command center and babysit AI workers. He compared it to a tmux grid. Yuchen Jin in the replies compared it to StarCraft.

These people are not joking. They're designing a future where "software engineer" means "guy who watches 12 AI agents on 6 monitors and makes sure none of them crashed."

And then.. AND THEN.. he casually says you'll be able to fork entire companies. Not code. Companies. Copy-paste someone's whole operation like it's an open source repo..

The man who mass produced software engineers just mass produced CEOs. And he did it in a tweet.

If you spent the last 5 years grinding LeetCode and building a résumé to get into FAANG, I genuinely don't know what to tell you right now.

The guy who wrote the playbook just burned it.

[Andrej Karpathy @karpathy]

All of these patterns as an example are just matters of “org code”. The IDE helps you build, run, manage them. You can’t fork classical orgs (eg Microsoft) but you’ll be able to fork agentic orgs.

[Suryansh Tiwari @Suryanshti777]

Most people treat CLAUDE.md like a prompt file.

That’s the mistake.

If you want Claude Code to feel like a senior engineer living inside your repo, your project needs structure.

Claude needs 4 things at all times:

• the why → what the system does
• the map → where things live
• the rules → what’s allowed / not allowed
• the workflows → how work gets done

I call this:

The Anatomy of a Claude Code Project 👇

━━━━━━━━━━━━━━━

1️⃣ CLAUDE.md = Repo Memory (keep it short)

This is the north star file.

Not a knowledge dump. Just:

• Purpose (WHY)
• Repo map (WHAT)
• Rules + commands (HOW)

If it gets too long, the model starts missing important context.

━━━━━━━━━━━━━━━

2️⃣ .claude/skills/ = Reusable Expert Modes

Stop rewriting instructions.

Turn common workflows into skills:

• code review checklist
• refactor playbook
• release procedure
• debugging flow

Result:
Consistency across sessions and teammates.

━━━━━━━━━━━━━━━

3️⃣ .claude/hooks/ = Guardrails

Models forget.

Hooks don’t.

Use them for things that must be deterministic:

• run formatter after edits
• run tests on core changes
• block unsafe directories (auth, billing, migrations)

━━━━━━━━━━━━━━━

4️⃣ docs/ = Progressive Context

Don’t bloat prompts.

Claude just needs to know where truth lives:

• architecture overview
• ADRs (engineering decisions)
• operational runbooks

━━━━━━━━━━━━━━━

5️⃣ Local CLAUDE.md for risky modules

Put small files near sharp edges:

src/auth/CLAUDE.md
src/persistence/CLAUDE.md
infra/CLAUDE.md

Now Claude sees the gotchas exactly when it works there.

━━━━━━━━━━━━━━━

Prompting is temporary.

Structure is permanent.

When your repo is organized this way, Claude stops behaving like a chatbot…

…and starts acting like a project-native engineer.

[Ernesto Lopez @ErnestoSOFTWARE]

Don’t waste your time..

this is literally the only prompt you need to clone any design Into your app.

[Chamath Palihapitiya @chamath]

As coding agents start compiling English into code, the definition of WHAT to build becomes the highest leverage work.

That means teams need to collaborate deeply on requirements.

Some people write specs as markdown in Git repositories. But that isn’t a collaborative environment. PMs aren’t living in IDEs or PRs.

Software Factory brings a Cursor-like agent experience into a multiplayer document editor.

If you’ve wondered why Google Docs never built a true, agent-driven editor: it’s extremely hard.

We just shipped one.

Try it here: https://t.co/fkfTXgcI8c

[Vadim @VadimStrizheus]

This is one of the best Claude Cowork articles on X.

Must read for Claude users. 👇

[Rohan Paul @rohanpaul_ai]

The paper says the best way to manage AI context is to treat everything like a file system.

Today, a model's knowledge sits in separate prompts, databases, tools, and logs, so context engineering pulls this into a coherent system.

The paper proposes an agentic file system where every memory, tool, external source, and human note appears as a file in a shared space.

A persistent context repository separates raw history, long term memory, and short lived scratchpads, so the model's prompt holds only the slice needed right now.

Every access and transformation is logged with timestamps and provenance, giving a trail for how information, tools, and human feedback shaped an answer.

Because large language models see only limited context each call and forget past ones, the architecture adds a constructor to shrink context, an updater to swap pieces, and an evaluator to check answers and update memory.

All of this is implemented in the AIGNE framework, where agents remember past conversations and call services like GitHub through the same file style interface, turning scattered prompts into a reusable context layer.

----

Paper Link – arxiv. org/abs/2512.05470

Paper Title: "Everything is Context: Agentic File System Abstraction for Context Engineering"

[Harshil Tomar @Hartdrawss]

25 signs your vibe-coded app is a TICKING BOMB !

1. API keys hardcoded “for now”
2. No /health endpoint, you just hit the homepage
3. Schema changes live in your head, not migrations
4. Every query is SELECT * and vibes
5. Error handling = console.log(e) and hope
6. No rate limit on auth or writes
7. UTC, local time, and “JS default” all mixed
8. README is empty or wrong
9. No staging env, just “dev” and “prod-ish”
10. One god component owns the whole screen
11. No analytics, just “feels like people use it”
12. You say “we’ll clean this up after launch” every week
13. Env vars live only on your laptop, nowhere else documented
14. Frontend talks directly to 5 different third-party APIs with no wrapper
15. No monitoring or alerts – you find out it’s down from a DM
16. Logs only exist in your local terminal history
17. DB backups are “automatic”… but you’ve never tested a restore
18. Feature flags = commenting code in and out
19. Deploys are done from your local machine with one random script
20. No input validation, you trust whatever the client sends
21. CORS is set to * because “it fixed the error”
22. CI is “I ran it once locally and it worked”
23. Same API token reused across staging, prod, and local
24. Only one person actually knows how to run or deploy the app

Bookmark this to defuse today LOL

[Michael Truell @mntruell]

https://t.co/jZKOk8RAsV