THESE GUYS ARE COOKING WE USE THIS AND ITS PHENOMENAL
this guy is the best engineer i know
Everyone’s shopping for agent sandboxes. Few have a dev env an agent can actually start.
Put your agent in the best sandbox there is, and it still hangs on “Are you sure? [y/N]”.
How I fixed ours:
dev-env-for-agents.pastehtml…
Alsufi retweeted
i met a girl at an event who was looking for growth roles at a startup and asked for my advice.
i told her to start building her personal brand on x
building a company brand is hard, especially in a boring industry (like sandboxes...)
you’re a person with opinions, stories, and a face. a company has to work much harder to get anyone to care.
if you can’t get people interested in what you have to say, how are you going to get them interested in a company?
Introducing Cortad Simulations.
→ Install Cortad directly in Claude Code, Cursor, or Codex.
→ Your coding agent runs simulations against your actual app, on your machine, using your own model.
→ Hundreds of realistic conversations test your agent's behavior across prompts, tools, memory, and real user journeys.
→ Every response is evaluated against up to 100 behavioral questions.
→ Cortad returns the failures, exact conversations, and code locations to your coding agent.
→ Your agent fixes the problems and reruns the failing conversations to verify the improvements.
No evals to write. No test cases to maintain.
Your coding agent builds. Cortad makes sure it actually works.
You can literally turn any data into one coherent base of context for your agents!
We (actually) KILLED Palantir this time!!!
New updates, better ontology generation, cloud service, and smarter agents!
You can deploy it locally here: github.com/HelixDB/helix-fou…
or use the cloud: foundry.helix-db.com
Alsufi retweeted
Excited to announce that Cortad has decided to begin their pre seed raise, with a valuation cap of $30.
Any connects to VC partners, investors, Peter Thiel, Kanye West, or Theo Von would be much appreciated.












