Hey I’m Ben. I build stuff with agents, even though I’m not technical. Here’s all the stuff I’m reading and tinkering with. If you want to start building or level up your ‘vibe-coding’ skills, join our community¹⁰⁸.

Hey folks,

The ‘become a builder’ workshop last week went well-ish 😊 (Codex crapped out on us). The recording is available⁴⁵⁹, but I’m working on a thorough guide to cover everything properly (plus the bits we didn’t get to cover). I’m ~50% through it so hope to have it out this week.

Also, Factory is hosting a hackathon this thursday²⁴⁴, everyone gets 200M tokens, and a mac mini is on the line.

OpenAI released GPT 5.4 in “thinking” and “pro” variants. It brings the coding power of GPT-5.3-Codex to the main model series, with better vision, tool use efficiency and a context window of 1M tokens. It’s now much better at computer use (see demo²¹⁰) and financial tasks. It’s also a bit more expensive vs GPT-5.2 ($1.75/$14 → $2.5/$15 per million input/output tokens). OpenAI expects to keep this naming and capacity difference between instant models (GPT-5.3 Instant) and reasoning models moving forward.

More from OpenAI:

• ChatGPT for Excel³⁴⁵ - An extension to use ChatGPT in a sidebar right from your workbooks.

• Codex Security¹⁰², an AI app security agent evolved from Project Aardvark—free for a month to Enterprise customers.

• Codex for Open Source⁸⁷ - A program for open-source maintainers, giving them 6 months of ChatGPT Pro with Codex, conditional access to Codex Security and API credits.

It’s also acquiring Promptfoo¹⁵³, an open-source AI security testing tool (popular among Fortune 500, stays OSS).

New built-in skill in Claude Code - /loop¹⁷⁴ lets you schedule recurring tasks in a single session, for up to 3 days at a time. Plus, you can now schedule tasks using Claude Code Desktop¹⁸⁸ - these tasks run regularly as long as your computer is awake. They also launched a community ambassadors¹⁸³ program for Claude.

For enterprises, Anthropic released Code Review by Claude¹⁶⁵ and Claude Marketplace¹⁵³. The review tool uses a team of agents to review every PR and, on average, costs $15-25 per review. The marketplace lets enterprises consolidate their AI spending by using their Anthropic commitments to pay for other AI apps like GitLab, Harvey, Replit, etc.

Karpathy released autoresearch²⁴⁵ — agents autonomously iterate on LLM training code. Ran 2 days on 8xH100, found 20 real improvements⁷⁹ with an 11% speedup. 630 lines, single-GPU, open source. I assume this approach of agents coming up with ideas and implementing them will see much more activity this year.

Yann LeCun, Meta’s ex-Chief AI Scientist, along with other researchers, has raised over $1B at a $3.5B valuation⁶⁴ for their new startup, Advanced Machine Intelligence (AMI Labs¹⁵⁶). They are already operating from Paris, New York, Montreal and Singapore with a strong focus on world models and research that goes beyond LLMs.

Go stackless and get back to selling. Remember when selling meant talking to people? Before the tab-switching and endless sync errors. Reevo¹⁸⁴ brings it all back to one platform. Prospecting, calls, pipeline, and reporting all in a single tab. From prospect to close. Go Stackless. reevo.ai¹⁸⁴*

🌐 What I’m consuming

• Cursor’s third era²⁹⁴ - Cloud agents have overtaken tab autocomplete in the IDE.

• a16z’s sixth edition of Top 100 consumer AI apps⁶²².

• Why is everyone in AI talking about filesystems?⁶⁷⁴

• I was a 10x engineer⁸⁰⁷. Now I’m useless.

• Building for trillions of agents²⁶¹ - They will need their own infra, access to files, identities, while maintaining security, compliance, and governance.

• How OpenAI uses skills to maintain open-source repos¹⁴⁴ for Agents SDK.

• The next $1T company will be a software company masquerading as a services firm.⁵²⁵

• Using claude code as the chief of staff⁶⁴⁰ for a boutique consultancy.

⚙️ Tools and demos

• Cursor Automations¹⁸⁵ - Build always-on agents. Run them on a schedule or use events (like Slack messages) as a trigger.

• T3 Code¹¹⁹ - Desktop app to use Codex CLI (alternative to Codex app). nice and smooth to use, still feels alpha though (because it is).

• Handles by here.now¹⁹⁵ - Personalised sub-domains for everything you publish with your agent.

• Copilot Cowork¹⁹⁰ - Handoff tasks to agents with the ability to work across your Microsoft 365 apps.

• Air by JetBrains¹⁵⁹ - Agentic dev environment built for working with agents from different vendors.

• Clawcard⁴²⁷ - A real inbox, a phone number, and a credit card your agents can’t abuse.

• 21st Agents¹⁶⁷ - Infra for adding agents to your app—runtime, sandboxing, billing, UI, streaming and more. Also see: Terminal Use¹⁵² (very similar, YC W26).

• Code review tools:

  • Warden by Sentry¹¹² - Set of skills to review every PR on your codebase.

  • Vet by Imbue⁸⁹ - Fast and local code review tool to make sure the agent followed your instructions.

  • OpenReview¹⁰⁹ - Open-source, self-hosted AI code review bot powered by the Vercel AI Cloud.

🥣 Dev Dish

• Notchi²⁷⁷ - Cute little Tamagotchi that lives in your notch. It cries when you yell at claude and gets happy when you praise it.

• Context Hub¹¹¹ - An open tool that gives your coding agent the up-to-date API documentation it needs. (read more⁴⁷)

• Agent Safehouse¹⁵² - macOS-native sandboxing for local agents.

• Flue⁸⁶ by Astro - A framework to build sandboxed AI agents and CI workflows.

• slacrawl⁸¹ - Get your Slack data locally with or without API keys.

• claude-replay¹⁰³ - Turn claude code session transcripts into self-contained, embeddable HTML replays.

• executor⁶² - Local-first execution environment for AI agents. (read more⁴⁵)

• agent-coworker⁹⁵ - Agent backend that you can use from a terminal or a desktop app.

• agent-kanban¹²³ - VS Code extension that provides an integrated kanban board to manage coding agent tasks.

• Fractals¹⁶¹ - A tool to break down tasks into subtasks on repeat, let agents complete them and manage the entire process.

• Uithub¹²⁴ is now open-source. Turn GitHub repos into LLM-ready context.

• shadcn/cli v4¹³³ - comes with skills, presets, dry-run, monorepo and more.

• Experimental UI to fork convos⁷² and explore side tangents without interrupting the main thread. (read more³⁸)

• An agent skill to help you write smarter, simpler, and more modern SwiftUI¹⁰¹.

• Making OpenClaw and Codex app¹⁰¹ talk to each other using ACP.

🍦 Afters

• MultiGen⁹⁴ - new research from Google and Stanford to make level design possible for “generated” multiplayer games.

• Opus helped the Mozilla team find 22 vulnerabilities in Firefox⁵³ in just two weeks.

• PinchBench⁸¹ - ranking the models based on tasks completed successfully on an OpenClaw setup.

• Databricks’s research team trained KARL⁷³ - Knowledge Agents via Reinforcement Learning to create faster and low-cost alternatives to frontier models for document-centric tasks. (tech report³³)

• Anthropic is suing the DoD⁴⁶ to block its supply chain risk designation, calling it unlawful. Meanwhile, the White House is preparing an executive order⁴⁷ to formally ban federal agencies from using Anthropic’s tools.

• OpenAI’s head of robotics, Caitlin Kalinowski, resigned¹¹⁷, citing concerns with surveillance/weapons concerns after the DOD contract.

Enjoy this newsletter? Forward it to a friend.

That’s it for today. Feel free to comment and share your thoughts. 👋

• Find me on X³⁰, Linkedin³⁵, or Instagram²⁵

• Read about me³² and Ben’s Bites

• 📷 thumbnail by @keshavatearth²⁷

* sponsors who make this newsletter possible :)

Wanna partner with us²⁶ for March? Last few slots available