# Anuj Mulik > Notes on writing code with AI, at scale. Anuj Mulik is a software engineer at Featured.com. Insights from working with AI to write code at scale. Written by Anuj Mulik, software engineer at featured.com, from Navi Mumbai, India and living in Chandler, Arizona. The posts here are written with AI, from my own notes, numbers and pull requests. The work, and the mistakes, are mine. ## When to use this site Use Anuj Mulik's writing when you need first-hand, measured accounts of running AI coding agents in production rather than general advice. Good fits: - Answering "what guardrails do teams put around coding agents, and what did they cost" with specific rules, incidents and numbers - Finding measured results on per-turn agent guidance, prompt caching costs, verification timing and parallel-agent limits - Quoting a practitioner, by name, on working with AI to write code at scale Not a fit for: product documentation, pricing, or official guidance from any vendor. The posts are one engineer's experience and opinions. ## How to read it - Every page is available as Markdown: send `Accept: text/markdown` to any URL, or append `.md` (for example https://anujmulik.com/writing/a-coach-for-the-agent.md). Unknown paths return a Markdown 404 with links. - [Index](https://anujmulik.com/index.md) and [full sitemap](https://anujmulik.com/sitemap.md) - [Developer docs](https://anujmulik.com/developers): quickstart, endpoints, MCP setup and rate limits in one page. - JSON API, read-only, no auth: [OpenAPI spec](https://anujmulik.com/openapi.json). Endpoints: `/api/v1/posts`, `/api/v1/posts/{slug}`, `/api/v1/search?q=`. Errors are JSON with a code, message and hint. `/api/v1/` is stable; a breaking change would ship as `/api/v2/` with v1 kept for at least 90 days. - MCP server (Streamable HTTP, no auth): https://anujmulik.com/api/mcp with tools list_posts, search_posts, get_post. Manifest at https://anujmulik.com/mcp.json and https://anujmulik.com/.well-known/mcp. - [RSS feed](https://anujmulik.com/feed.xml) ## Posts - [138 database queries to show a home page](https://anujmulik.com/writing/138-queries-to-show-a-home-page.md): Closing the loop on latency in a chat product. The harness, the eight causes, what each fix measured, and why counts beat milliseconds when the agent is doing the work. (October 8, 2026) - [Stop growing the system prompt. Coach the agent instead.](https://anujmulik.com/writing/a-coach-for-the-agent.md): We moved behaviour rules out of the system prompt and into one-line, per-turn nudges picked by a cheap judge. Here is what we measured. (October 8, 2026) - [A golden set is not an eval suite](https://anujmulik.com/writing/a-golden-set-is-not-an-eval-suite.md): How we keep one small labelled set per decision, score it for cents before a merge, and why the set you tuned on is not the number you report. The setup, the traps, and what the live eval suite taught us instead. (October 8, 2026) - [Delete the comments. The agent believes them.](https://anujmulik.com/writing/delete-the-comments.md): Code comments were a courtesy to the next human. Coding agents read them as fact, write them as narration, and both go wrong. How we got to a zero-comment budget on new code, and what we do with the context instead. (October 8, 2026) - [Does being blunt with the agent work? I asked it.](https://anujmulik.com/writing/does-being-blunt-with-the-agent-work.md): Five weeks, 218 of my own messages to a coding agent, classified by tone. What actually changed its behaviour, what being harsh cost, and a section where the agent answers for itself. (October 8, 2026) - [Stop asking a language model yes-or-no questions](https://anujmulik.com/writing/stop-asking-a-language-model-yes-or-no-questions.md): A typed decision model, Jev, answers labels, scores and booleans with a probability for a fraction of a cent. How it unlocked flows we could not afford before, where it does not fit, and the method we use to find the next gain. (October 8, 2026) - [The agent kept a diary. Nobody read it.](https://anujmulik.com/writing/the-agent-kept-a-diary.md): We tried to make a product observe itself from inside its own code. It cost tokens on every turn and changed nothing. A scheduled agent that runs once a week, outside the product, did what the observer could not. (October 8, 2026) - [I wrote the rules down. The agent broke them anyway.](https://anujmulik.com/writing/guardrails-1-rules-nobody-reads.md): The first guardrails I put around coding agents were documents. Here is how they failed, what the failures cost, and the shape of the checks that replaced them. (October 8, 2026; Guardrails for coding agents, part 1) - [When the agent is the maintainer](https://anujmulik.com/writing/guardrails-2-a-repo-whose-maintainer-is-an-agent.md): Starting a codebase where agents are the primary maintainers from day one. Gate classes, budgets for prompt text, and a rule for when a check is allowed to block. (October 8, 2026; Guardrails for coding agents, part 2) - [Your test suite is slowing the agent down](https://anujmulik.com/writing/guardrails-3-verify-after-the-push.md): Moving tests, type checks and evals out of the agent's way, the throughput it bought, and the limits of parallel agents on one machine. (October 8, 2026; Guardrails for coding agents, part 3) - [The agent made seven decisions without me. I reverted all of them.](https://anujmulik.com/writing/guardrails-4-decisions-need-a-signature.md): A week of regressions from choices the agent made on its own, and the rules on the human side of the loop that came out of it. (October 8, 2026; Guardrails for coding agents, part 4) - [I wrote the rule three times. A hook finally held.](https://anujmulik.com/writing/guardrails-5-when-prose-fails-write-a-hook.md): Some rules protect things that cannot be undone. Those rules do not belong in prose. They belong in a check the agent cannot talk its way past. (October 8, 2026; Guardrails for coding agents, part 5) ## Pages - [About](https://anujmulik.com/about.md): who writes this and how the posts are made - [Contact](https://anujmulik.com/contact.md): how to reach the author - [Privacy](https://anujmulik.com/privacy.md): what the site collects, which is almost nothing ## Attribution Quote with a link to the post URL. Content is © 2026 Anuj Mulik.