At 17:23 UTC on 2026-10-25, TypeScript educator Matt Pocock (@mattpocockuk, author of Total TypeScript and AI Hero) posted (post) that the next great "language for agents" should be as fast as Rust, have as little design space as Go, and have the strictest type/verification system ever written, calling Bend "an interesting first effort". When we pulled it, the post had about 3,130 likes, 84 reposts, 264 replies, 786 bookmarks and 220K views, and it reached X's trending stories. Mojo creator Chris Lattner replied (reply) recommending Mojo and suggesting "can target all hardware" be added to the list. This is an opinion debate, not a verified conclusion.
Key Takeaways
- ✓Main post: post, published 17:23 UTC on 10-25, with about 3,130 likes, 84 reposts, 264 replies, 786 bookmarks and 220K views when we pulled it
- ✓Three criteria: Rust-level speed, a Go-sized design space (few ways to write the same thing), and the strictest type/verification system ever written, so the compiler rather than a human catches agent mistakes
- ✓Chris Lattner replied recommending Mojo, citing advanced type features such as linear types and symbolic constraints, and suggested adding "can target all hardware" (reply, about 40 likes when pulled)
- ✓Bend creator Victor Taelin (@VictorTaelin) also joined the thread, saying Bend blocking an agent from breaking an app is "working as intended"
- ✓No benchmarks or controlled comparisons were shared; whether Bend or Mojo suits agent coding remains personal opinion
Key Decision Metrics at a Glance
Turn your technical choice into a development budget
Compare 40 dev plans & simulate token costs vs $20/mo subscriptions
Project Links & Resources
Direct AccessIn-Depth Technical Analysis
Background and the problem
Coding agents keep getting faster at writing code, so the bottleneck has moved to how humans can safely merge what they write. Other trending X posts in the same window, such as "if you're still merging most PRs by hand, you're behind" and "no human codes better than LLMs", share the same worry: review and verification are not keeping up with generation. At 17:23 UTC on 10-25, TypeScript educator Matt Pocock (@mattpocockuk) posted (post) that the next great "language for agents" should be as fast as Rust, have as little design space as Go, and have the strictest type/verification system ever written, calling Bend "an interesting first effort". When we pulled it, the post had about 3,130 likes, 264 replies and 220K views, and it reached X's trending stories.
How it works
The logic behind the three criteria is that agents are not hurt by writing a lot, but by writing wrong code that nobody notices. A small design space means fewer ways to express the same requirement, so model output is more consistent and easier to review. The strictest type and verification system hands part of code review to the compiler and rejects changes that break constraints before they run. High performance keeps safety from costing speed. Bend is an open-source high-level parallel language from HigherOrderCO (GitHub); its creator Victor Taelin said in the thread that Bend blocking an agent from breaking an app is "working as intended". Mojo creator Chris Lattner replied (reply) recommending Mojo, citing advanced type features such as linear types and symbolic constraints, and suggested adding "can target all hardware" to the list.
Benchmarks and measured results
This is an opinion debate on X. There are no public benchmarks, and nobody shared a controlled comparison of agents writing code in different languages. The only numbers we can check are engagement: the main post had about 3,130 likes, 84 reposts, 264 replies and 786 bookmarks, and Lattner's reply about 40 likes. The claim that Bend or Mojo suits agents better is, for now, personal opinion.
What developers can do now
You do not need to switch languages to apply the idea. In TypeScript, turn on strict mode and cut down on any; in Rust or Go projects, wire lint, type checks and tests into every agent loop so failures go straight back to the agent to fix; and agree on one way of writing things in AGENTS.md to shrink the design space by convention. If you want to try the new languages, look at the Bend repository and the Mojo site, start with small tasks, and watch the agent's error rate and number of fix rounds under a strict language before investing more.
Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.
Discussion & Comments
0Sign in to join the discussion
Connect with AI developers to exchange benchmark insights.