https://timurista.ai/writing/ Field notes on managing AI-assisted software delivery: how I scope the work, what I check before anything ships, and the bugs that got through anyway. en-us https://timurista.ai/writing/the-evaluator-forgot-the-session/ https://timurista.ai/writing/the-evaluator-forgot-the-session/ Fri, 13 Nov 2026 00:00:00 GMT A saved-state resume exposed a fresh-start assumption in a Socratic Tutor transcript checker. A local regression fix passed 32 synthetic tests, with state validation and tutoring quality explicitly outside its claims. https://timurista.ai/writing/the-rollback-that-made-the-new-column-lie/ https://timurista.ai/writing/the-rollback-that-made-the-new-column-lie/ Thu, 12 Nov 2026 00:00:00 GMT An old writer can remain compatible with an expanded schema while leaving new-column reads stale. A synthetic, serial, four-row PostgreSQL 16.15 test observed the rollback window: one legacy update and one legacy insert left a new-column-only read wrong on 2 of 4 rows, while the old-column-authoritative read stayed correct on 4 of 4 in this run. https://timurista.ai/writing/can-you-still-finish-when-the-assistant-disappears/ https://timurista.ai/writing/can-you-still-finish-when-the-assistant-disappears/ Fri, 06 Nov 2026 00:00:00 GMT A structured handoff note for paused AI-assisted work, a stale-claim check, and a proposed synthetic test. No participant study has been run. https://timurista.ai/writing/juniors-without-tickets/ https://timurista.ai/writing/juniors-without-tickets/ Thu, 05 Nov 2026 00:00:00 GMT The well-specified ticket was how junior engineers learned, and agents now absorb it. What the research says about learning with an AI assistant, why answer-first beats reveal-first, and the starter work I'd give a junior instead, with the gates I built into my own study tools as the worked example. https://timurista.ai/writing/any-app-can-rewrite-your-agents-rules/ https://timurista.ai/writing/any-app-can-rewrite-your-agents-rules/ Wed, 04 Nov 2026 00:00:00 GMT Opening Postman's desktop app turned on Claude Code and Codex plugins and wrote a startup hook, without asking. I read the issue, checked my own machine, and built a small open-source tool that notices when anything changes what your coding agents do. https://timurista.ai/writing/architect-of-intent/ https://timurista.ai/writing/architect-of-intent/ Sat, 31 Oct 2026 00:00:00 GMT A viral preprint says software engineering is ending and the human becomes an intent architect. I read it: one author, a title softened in six days, and one benchmark number that explains the job better than the headline. What intent architecture looks like in a real repo, and what it changes downstream. https://timurista.ai/writing/hand-the-next-owner-a-broken-run/ https://timurista.ai/writing/hand-the-next-owner-a-broken-run/ Mon, 26 Oct 2026 00:00:00 GMT Reassigning code owners shows who is responsible, not whether they can recover anything. A departure drill: give the next steward one known failed run, no access to the previous owner, and record where it stalls. A synthetic version with six levels of guidance, and what a pass can and can't prove. https://timurista.ai/writing/the-row-that-never-reached-the-comparison/ https://timurista.ai/writing/the-row-that-never-reached-the-comparison/ Fri, 23 Oct 2026 00:00:00 GMT A dual-write migration check that compares rows present in both stores can report a clean match while an order is missing entirely. A synthetic two-store demo with a real crash between commits, the audit that catches it, and a reconciler that repairs only from evidence. https://timurista.ai/writing/what-happens-after-you-press-stop/ https://timurista.ai/writing/what-happens-after-you-press-stop/ Mon, 19 Oct 2026 00:00:00 GMT A Stop button that ends in one word hides what actually happened. A small synthetic demo of a return summary: one confirmed write, one unknown, one never sent, the actions each one allows, and why a timeout must never turn unknown into failed. https://timurista.ai/writing/overtrain-review-loops/ https://timurista.ai/writing/overtrain-review-loops/ Sat, 17 Oct 2026 00:00:00 GMT Ten days of Read The Games: 11-a-side matches, a decision trainer, game review on the phone, motion capture, and boxing and sword fights in the headset. The part worth copying is how I checked the AI against reality: replayable seeded games, annotated reviews, and tracked match footage, each one closing a loop with a number. https://timurista.ai/writing/overtrain-mixed-reality-week/ https://timurista.ai/writing/overtrain-mixed-reality-week/ Thu, 08 Oct 2026 00:00:00 GMT A week building Overtrain, a mixed-reality soccer trainer for Meta Quest, plus a desktop mode and a phone app from the same code. Most of the hard work was spatial: getting the play area to sit on the real floor, inside the boundary you drew, and stay there. Here's what broke, how each fix worked, and what I'd do differently. https://timurista.ai/writing/when-the-agent-finishes-yesterdays-task/ https://timurista.ai/writing/when-the-agent-finishes-yesterdays-task/ Wed, 07 Oct 2026 00:00:00 GMT What should a product do when an async agent's result arrives after the user changed or cancelled the goal? A deterministic demo with 11 synthetic schedules and three admission policies, the race a pre-check misses, and the wording for current, pending, cancelled and conflict states. https://timurista.ai/writing/outrider-three-js-wasm-renderer/ https://timurista.ai/writing/outrider-three-js-wasm-renderer/ Fri, 02 Oct 2026 00:00:00 GMT Outrider is a Meta Quest game where your indoor bike is the controller. Where to find its best moments, and what I learned when three.js spent too much CPU on draw calls: I wrote a lean draw loop in TypeScript and again in Rust/WASM, benchmarked all three, and the language mattered less than the leanness. https://timurista.ai/writing/trendvesting-debugging-a-rebalance-loop/ https://timurista.ai/writing/trendvesting-debugging-a-rebalance-loop/ Fri, 25 Sep 2026 00:00:00 GMT A live ETF agent sold 9 shares of SGOV and bought them back once a minute, 183 times in a day. How I found the loop, the four bugs behind it, the fixes and tests, and the diagnostics call that now catches all of them. https://timurista.ai/writing/unalloc-who-pays-for-the-kv-cache/ https://timurista.ai/writing/unalloc-who-pays-for-the-kv-cache/ Tue, 15 Sep 2026 00:00:00 GMT A guide for platform and FinOps leads who have to split shared AI spend: which meter to charge by, where idle capacity goes, how to give every dollar one path into the ledger, and what a joined ledger can answer. Backed by measurements from unalloc, my open-source cost-attribution tool. https://timurista.ai/writing/timurista-ai-rebuilding-this-website/ https://timurista.ai/writing/timurista-ai-rebuilding-this-website/ Mon, 14 Sep 2026 00:00:00 GMT How I scoped, reviewed and shipped the rebuild of this site: a framing I pulled back within the hour, numbers I deleted because I couldn't source them, a form that showed my phone number to the wrong people, and a retired Worker route that quietly shadowed the new API. https://timurista.ai/writing/unalloc-validating-ai-cost-attribution/ https://timurista.ai/writing/unalloc-validating-ai-cost-attribution/ Mon, 14 Sep 2026 00:00:00 GMT How I scoped and checked unalloc, an open-source tool that joins Kubernetes and LLM provider bills: nine defects caught before release, a budget-capped H100 run with a verified teardown, a paper build that refuses silent typos, and a headline I softened because no meter is ground truth. https://timurista.ai/writing/tendform-managing-a-solo-saas-build/ https://timurista.ai/writing/tendform-managing-a-solo-saas-build/ Sun, 13 Sep 2026 00:00:00 GMT How I scoped, reviewed and validated a HIPAA-capable form builder I built alone: the rules I wrote down, the checks I trusted, the 14 bugs that got through anyway, and the process I'd change. https://timurista.ai/writing/tigermill-quality-gates-for-ai-content/ https://timurista.ai/writing/tigermill-quality-gates-for-ai-content/ Sun, 13 Sep 2026 00:00:00 GMT How I defined acceptance for a novel-to-webtoon pipeline: a gate after every stage, approvals that name who made them, spend behind a human trigger, and tests that guard the defects I already paid for once. https://timurista.ai/writing/trendvesting-validating-production-ai/ https://timurista.ai/writing/trendvesting-validating-production-ai/ Sun, 13 Sep 2026 00:00:00 GMT How I review and validate an AI trading system I own: change records with production numbers, invariant tests, reconciliation audits and calibration checks, and the numbers they overturned.