Dotu (@dotudev) header
Dotu (@dotudev) profile picture

I treat AI agents like my engineering team

Joined October 2026
Photo by Dotu (@dotudev)Photo by Dotu (@dotudev)Photo by Dotu (@dotudev)Photo by Dotu (@dotudev)Photo by Dotu (@dotudev)Photo by Dotu (@dotudev)Photo by Dotu (@dotudev)Photo by Dotu (@dotudev)Photo by Dotu (@dotudev)Photo by Dotu (@dotudev)
you asked for a one line fix the agent rewrote 14 files and "cleaned up" three folders you didn't mention what's your move?
19
Claude Code or Codex which one burns through your limit faster? 👀
30
Claude had two incidents yesterday. Logins broke for about 2 hours, then the API and Claude Code threw elevated errors for a couple more. The fallback flag only covers overload. Not login, not rate limits. I keep the plan in a file in the repo so switching agents mid-task isn't a restart.
1
28
What's the first thing you'd hand off to an AI agent if you could trust it 100%? Emails Coding Marketing Customer support Marketing for me. I'd build all day if someone else posted.
1
34
Developers, which one would you pay for if you could only keep one: Claude Code, Codex, or Cursor?
21
Quick bet for the Claude users here. One year after the IPO, what changes first? Pricing Rate limits Release speed Nothing I'm saying rate limits.
Our initial public offering: ipo.anthropic.com Legal disclaimer: Anthropic has filed a registration statement (including a prospectus) with the SEC for the offering to which this communication relates. Before you invest, you should read the prospectus in that registration statement (available at: sec.gov/Archives/edgar/data/…) and other documents Anthropic has filed with the SEC for more complete information about Anthropic and this offering. You may get these documents for free by visiting EDGAR on the SEC Web site at sec.gov. Alternatively, copies of the preliminary prospectus may be obtained from: Goldman Sachs & Co. LLC, Attention: Prospectus Department, 200 West Street, New York, NY 10282, by telephone at (866) 471-2526, or by email at [email protected]; or Morgan Stanley & Co. LLC, Attention: Prospectus Department, 180 Varick Street, 2nd Floor, New York, NY 10014. The securities may not be sold, nor may offers to buy be accepted, prior to the time the registration statement becomes effective. This post is not an offer to sell, or a solicitation of an offer to buy, the securities in any jurisdiction where the offer or sale is not permitted.
1
1
32
Where does your coding agent actually run? Terminal IDE Cloud Which one are you using right now?
1
17
A CLAUDE.md file asks Claude to behave. A hook enforces it. It runs a shell command at a fixed point in the run, whether the model remembers the rule or not.
2
1
196
The hook gets JSON on stdin with what Claude is about to do. It answers with an exit code. Exit 2 on a PreToolUse hook blocks the call and hands the reason back to Claude.
1
108
Start with these two. A PostToolUse hook that formats every file Claude edits. A PreToolUse hook on Bash that blocks rm -rf before it runs. What would you block first?
103
If the same instructions keep going into Claude Code, stop pasting them. Write the block down once as a skill and let Claude load it when the job matches.
1
1
209
Personal skills live in ~/.claude/skills/ and follow you across projects. Project skills live in .claude/skills/ and travel with the repo. Run it with /skill-name or let Claude pick it up from the description.
1
87
Skills that touch real things get one extra line. disable-model-invocation: true stops Claude loading it on its own. A deploy skill only runs when you type its name. What is the first instruction you would never paste again?
77
Would your shipped code survive a check that deep?
Announcing a major milestone achieved by my team @GoogleDeepMind: we have proven key security properties of the machine code underlying the Signal messaging protocol. This is the first time the actual production code of a widely deployed protocol has been formally verified – by any method, with or without AI – and at this depth. More than a research breakthrough, it’s a prime example of how frontier AI can be a force multiplier to counter the speed and scale of modern cyber threats. Formal verification offers a higher standard of security; however, it’s extremely complex and time consuming, which is why in many cases it’s considered to be impractical. Similar efforts could take teams decades of combined effort. We did it in a month’s time. Using AlphaProof Code, Google DeepMind's new agentic framework for formal verification that runs on LLMs including Gemini Carbon, we verified the Signal protocol down to the actual code running on devices.  This milestone lays the groundwork for a future in which vital systems, from medical device firmware to power grid software, ship with mathematical proof that their protections work. Reaching that future will take continued research from our community. In that spirit, we're open-sourcing our formal specifications and verification artifacts so others can build on this work. On a personal note, this moment is of deep significance to me. I started my own research journey more than two decades ago as an intern working on checking the consistency of complex software systems. Seeing this grand challenge made practical with modern AI is extraordinary. I'm extremely proud of my team members: Swarat Chaudhuri, Andrew Ferraiuolo, Anton Kovsharov, Tej Chajed, Sam Schlesinger,  Karn Seth, Jonathan Katz, and Charles Sutton for taking on this grand challenge and delivering, and grateful to the Signal team for their partnership. deepmind.google/blog/formal-…
107
Which breaks first in your app, the state machine someone wrote down or the one nobody did? Would you draw that one first with /statechart?
try this next “use /create-formal-verification-skill to model these state machines, one skill per subsystem. and use /statechart to visualize my most important user flows” both skills are in pstack github.com/cursor/plugins/tr…
154
A subagent only earns its keep on a job that stands alone. Hand it one that needs the main chat and it just burns context.
1
304
Keep it in when step two needs step one. Two agents on the same file step on each other. A five-minute job costs more to hand off than to do.
1
129
One file, one writer, one outcome per handoff. Name the files up front. What do you hand off first?
129