Hacker News: bharxhav

New comment by bharxhav in "Learnings from 100K lines of Rust with AI (2025)"

bharxhav — Wed, 20 May 2026 11:22:27 +0000

Rust is about abstractions more than code. You can ask AI to "Optimize/Test/Clarify" but at the end of the day you should be willing to blindly agree to it's output or spend more time reviewing someone else's code.

New comment by bharxhav in "A Streaming First Language for Generative UI"

bharxhav — Thu, 14 May 2026 10:50:44 +0000

Didn't Vercel ship something like this last year?

https://github.com/vercel-labs/json-render

New comment by bharxhav in "AI Search Visibility: The Practical Guide to Generative Engine Optimization"

bharxhav — Thu, 14 May 2026 10:40:15 +0000

I feel it's far simpler if you optimize for:

()

New comment by bharxhav in "Claude subscriptions no longer include Agent SDK and Claude -p usage"

bharxhav — Thu, 14 May 2026 10:35:13 +0000

This is very generous. The agent sdk and -p were abused by other devs and "startups." They tried to block them, and we had the Hermes gate disaster. This is a very clean and low maintenance way to own their moat back.

Anthropic is selling subsidized products and not a subsidized AI, and they have all rights to.

New comment by bharxhav in "Beware AI Productivity Theater"

bharxhav — Thu, 14 May 2026 10:28:41 +0000

Many of us did fall for it. The dopamine hit we get from working hard and shipping a PR is now "automated." This is like tiktok for devs.

New comment by bharxhav in "Rewrite Bun in Rust has been merged"

bharxhav — Thu, 14 May 2026 10:22:44 +0000

This canary will never leave the mine. (unless Anthropic opens their wallet again)

New comment by bharxhav in "Agents for financial services and insurance"

bharxhav — Wed, 06 May 2026 13:20:09 +0000

"agents"

Damn, they keep brainwashing people into thinking LLMs are intelligent, meanwhile they be translating into somewhat coherent words and fool the world.

When they're right, they polish their intelligence badge. When they're wrong, we prompted incorrectly.

New comment by bharxhav in "Show HN: Red Squares – GitHub outages as contributions"

bharxhav — Wed, 06 May 2026 11:04:48 +0000

Would be interesting to see if this correlated with their release cycles.

New comment by bharxhav in "Further human + AI + proof assistant work on Knuth's "Claude Cycles" problem"

bharxhav — Sat, 28 Mar 2026 23:10:10 +0000

Ramanujan is a good analogy for this situation. Theories could be right/wrong, until there's a proof. Same with anything AI produces. There's always a "told you so" baked in with it's response.