sarthakaggarwal

3 ms·

1.
by sarthakaggarwal 7mo ago
The philosophical question here is fascinating — if an AI rewrites every line, is it still the same codebase? At what point does the Ship of Theseus argument apply to licensing? Practically though, I wonder how much this cost in API calls.
2.
by sarthakaggarwal 7mo ago
This resonates with AI coding tools. Claude Code doesn't know when to stop —it'll do 30 agentic loops for something that should have taken 3. The cost implications of a tool that doesn't know when it's done are brutal.
3.
by sarthakaggarwal 7mo ago
One thing I've noticed working with Claude Code extensively is that the quality is high but the cost predictability is near zero. Same prompt can cost $2 or $15 depending on how many agentic loops it takes. The capability is there, the
4.
by sarthakaggarwal 7mo ago
Interesting - I've been digging into Claude Code's behavior patterns too, but from the cost side. The variance in how many API calls a task takes is wild. A simple refactoring can be 3 calls or 30 depending on codebase state. Curi
5.
by sarthakaggarwal 7mo ago
Happy to answer any questions about the approach. The key insight is that conformal prediction gives you calibrated confidence intervals — if we target 80% coverage, we actually get 81% of actual costs landing within the predicted range. No
6.
Show HN: Tarmac – Know what Claude Code will cost before you run it(github.com)
2 points by sarthakaggarwal 7mo ago|1 comments
7.
Ask HN: How are you estimating API costs before committing to an architecture?
1 points by sarthakaggarwal 7mo ago|2 comments
8.
End-to-End RL Post-Training
4 points by sarthakaggarwal 1y ago|2 comments
9.
by sarthakaggarwal 4y ago
Agreed, I have a similar thought process. I've found it very hard to come up with an idea simply because I think that something like this already exists, even though I know I can build or at least try to build almost anything.