Finch

Siddharth Nanda - Microsoft Engineer Reveals how Engineering Will Never be the Same Again

"Software has been solved." Most of the best engineers I know have landed on this exact conclusion.

Last week I sat down with Siddharth Nanda, who went from writing every line of code by hand at Microsoft and Atlassian to shipping 30,000 lines a week with 0% of it written by him.

He's now at Finch, a startup that's raised $20M+ to automate admin work at personal injury law firms, where 100% of the code he writes comes from agents. He's seen big tech before AI and a startup fully running on it, so he has a pretty unique read on where this all goes. We get into why he hasn't written a line of code in 8 months, why middle management is getting hit hardest by the layoffs, and why the productivity studies saying companies are getting slower are right, but only for companies with the wrong people.

We also dig into the tooling itself: Codex vs Claude Code, what harness engineering actually is, running Devin agents in parallel, and why the cost of producing code is approaching zero, and what that means for the kind of engineer who thrives from here.

If you're building with agents, this one's full of hard-won takes from someone doing it every day.

Siddharth Nanda - Microsoft Engineer Reveals how Engineering Will Never be the Same Again
Episode still: Siddharth Nanda, Finch

Why this matters in practice

As agents produce more code, engineering leverage shifts toward decomposition, context, review, and judgment about what should ship.

Chapters

  1. 00:00Intro
  2. 01:16What engineering looked like at Microsoft before ChatGPT
  3. 03:08The first AI tools inside big tech (and how gated they were)
  4. 05:17Shipping 30,000 lines a week, none of it written by hand
  5. 06:08Why teams are flattening
  6. 08:26Why managers need to get back to writing code
  7. 11:51How the management layer actually changes
  8. 14:07What happens to PMs and designers
  9. 16:38Why everyone becomes a builder now
  10. 20:07How to actually learn agentic engineering
  11. 22:42Finding new tools on X before anyone else
  12. 23:54AI adoption is showing up in performance reviews
  13. 25:45Building a culture that shares tools daily
  14. 27:40The AI productivity paradox, explained
  15. 29:35Interviewing with agents instead of against them
  16. 31:00Why LeetCode is dead and fundamentals aren't
  17. 34:11Who wins and who loses from here
  18. 35:36Codex vs Claude Code: what he actually uses
  19. 36:06What harness engineering really means
  20. 38:21Evals and benchmarks that matter
  21. 39:59Desktop apps, Devin, and running agents in parallel
  22. 43:30Remote environments and working in monorepos
  23. 44:34Skills, plugins, and hooks
  24. 47:57Parallel worktrees and staying focused
  25. 50:42Agent teams vs subagents
  26. 53:06The Finch mission and wrap up

Related topics

More episodes