https://blog.turboawesome.win/categories/engineering/ Recent content in Engineering on Bits, Trades & Systems Hugo en-us Wed, 13 May 2026 09:00:00 +0000 https://blog.turboawesome.win/2026/05/a-software-engineer-plays-quant-building-a-market-data-research-stack-in-python/ Wed, 13 May 2026 09:00:00 +0000 https://blog.turboawesome.win/2026/05/a-software-engineer-plays-quant-building-a-market-data-research-stack-in-python/ I spent a decade building the systems trading firms run on, without ever being the person making the trading decisions. This is the first post in a series where I cross the floor — a software engineer applying engineering discipline to market-data analysis in Python with numpy and pandas. Starting with the unglamorous foundation everything else depends on. https://blog.turboawesome.win/2025/10/migrating-a-production-service-from-zap-to-slog-notes-from-the-trenches/ Tue, 07 Oct 2025 09:15:00 +0000 https://blog.turboawesome.win/2025/10/migrating-a-production-service-from-zap-to-slog-notes-from-the-trenches/ log/slog landed in the standard library in Go 1.21, but most production services are still on zap or zerolog. Here is what an actual migration looks like: where slog wins, where zap still wins, and how to move incrementally without a big-bang rewrite. https://blog.turboawesome.win/2025/08/context-engineering-what-the-term-actually-means-and-what-it-doesnt/ Tue, 19 Aug 2025 09:30:00 +0000 https://blog.turboawesome.win/2025/08/context-engineering-what-the-term-actually-means-and-what-it-doesnt/ "Context engineering" became the term of the year in 2025, mostly used to mean nothing. Here is what it actually refers to as an engineering discipline: managing the finite token budget of a model's context window as a resource, with the same rigour you'd apply to memory or a cache. https://blog.turboawesome.win/2025/06/go-1.23-range-over-functions-what-its-for-and-what-it-isnt/ Wed, 18 Jun 2025 09:00:00 +0000 https://blog.turboawesome.win/2025/06/go-1.23-range-over-functions-what-its-for-and-what-it-isnt/ Go 1.23 shipped range over functions — the ability to range over an iterator function rather than a slice, map, or channel. The feature is genuinely useful for a specific class of problems and genuinely misunderstood by engineers reaching for it in places where a slice works better. https://blog.turboawesome.win/2025/03/ai-native-development-what-it-actually-means-to-use-these-tools-well/ Wed, 05 Mar 2025 10:55:00 +0000 https://blog.turboawesome.win/2025/03/ai-native-development-what-it-actually-means-to-use-these-tools-well/ AI coding tools have changed the texture of software development. Not by writing code for you, but by changing what's worth doing yourself and what isn't. A practitioner's view of where the leverage actually is. https://blog.turboawesome.win/2025/01/building-with-ai-coding-tools-what-actually-changes-and-what-doesnt/ Wed, 22 Jan 2025 10:08:00 +0000 https://blog.turboawesome.win/2025/01/building-with-ai-coding-tools-what-actually-changes-and-what-doesnt/ A year into using AI coding assistants seriously: what they've changed about how I work, where they still fall short, and the habits that make the difference between AI as a demo and AI as a productivity multiplier. https://blog.turboawesome.win/2024/11/cross-team-technical-alignment-at-scale/ Wed, 20 Nov 2024 10:30:00 +0000 https://blog.turboawesome.win/2024/11/cross-team-technical-alignment-at-scale/ Getting multiple engineering teams to agree on technical direction is hard. The approaches that work — architecture forums, decision records, working groups — and why lightweight coordination beats heavy governance. https://blog.turboawesome.win/2024/08/writing-rfcs-for-wide-audiences/ Wed, 21 Aug 2024 09:45:00 +0000 https://blog.turboawesome.win/2024/08/writing-rfcs-for-wide-audiences/ RFCs that need sign-off from multiple teams or leadership require a different structure than internal team proposals. The writing changes when the audience includes people who don't share your context. https://blog.turboawesome.win/2023/11/eleven-years-in-a-retrospective-on-careers-choices-and-compounding-knowledge/ Wed, 15 Nov 2023 15:41:00 +0000 https://blog.turboawesome.win/2023/11/eleven-years-in-a-retrospective-on-careers-choices-and-compounding-knowledge/ Eleven years of software engineering across trading firms, financial institutions, startups, and large technology companies. What compounds, what doesn't, and the decisions I'd make differently. https://blog.turboawesome.win/2023/08/navigating-org-change-as-an-engineer/ Wed, 16 Aug 2023 11:20:00 +0000 https://blog.turboawesome.win/2023/08/navigating-org-change-as-an-engineer/ Reorganisations, leadership changes, and strategy shifts are facts of life in growing companies. The engineers who navigate them well share some common practices — and some common misconceptions about what's actually changing. https://blog.turboawesome.win/2023/01/writing-idiomatic-go/ Wed, 11 Jan 2023 10:00:00 +0000 https://blog.turboawesome.win/2023/01/writing-idiomatic-go/ Idiomatic Go is not just stylistic preference — it reflects how the language is designed to be used. The patterns that experienced Go engineers reach for, and why they work better than the alternatives. https://blog.turboawesome.win/2022/10/postmortems-as-a-learning-tool-structure-culture-and-follow-through/ Wed, 05 Oct 2022 11:27:00 +0000 https://blog.turboawesome.win/2022/10/postmortems-as-a-learning-tool-structure-culture-and-follow-through/ A postmortem that concludes with 'be more careful' has failed. The structure and culture of blameless postmortems, and why follow-through is where most teams fall short. https://blog.turboawesome.win/2022/07/on-call-culture-that-doesnt-burn-out-your-team/ Wed, 06 Jul 2022 09:55:00 +0000 https://blog.turboawesome.win/2022/07/on-call-culture-that-doesnt-burn-out-your-team/ On-call is a forcing function for reliability — when engineers are paged for their own systems, the feedback loop on quality is immediate. Done badly, it burns out your team. The practices that keep it sustainable. https://blog.turboawesome.win/2022/04/platform-engineering-is-a-product-problem-not-a-technology-problem/ Wed, 06 Apr 2022 11:39:00 +0000 https://blog.turboawesome.win/2022/04/platform-engineering-is-a-product-problem-not-a-technology-problem/ Two years into a platform engineering role at a mid-sized financial technology firm. The core insight: developer platforms succeed when they're treated as products and fail when they're treated as infrastructure projects. https://blog.turboawesome.win/2022/02/engineering-roadmaps-planning-for-uncertainty/ Wed, 16 Feb 2022 09:30:00 +0000 https://blog.turboawesome.win/2022/02/engineering-roadmaps-planning-for-uncertainty/ Engineering roadmaps are commitments made under uncertainty. The planning practices that keep them honest — time horizon discipline, explicit uncertainty, and the difference between a roadmap and a plan. https://blog.turboawesome.win/2022/01/embedding-in-go-composition-over-inheritance-done-right/ Wed, 05 Jan 2022 10:08:00 +0000 https://blog.turboawesome.win/2022/01/embedding-in-go-composition-over-inheritance-done-right/ Go has no inheritance, but it has embedding — a composition mechanism that promotes methods to the outer type. Understanding what embedding actually does (and what it doesn't) prevents the common mistakes. https://blog.turboawesome.win/2021/10/technical-debt-is-a-balance-sheet-item-not-a-moral-failing/ Wed, 06 Oct 2021 10:18:00 +0000 https://blog.turboawesome.win/2021/10/technical-debt-is-a-balance-sheet-item-not-a-moral-failing/ The language around technical debt is usually moral: 'bad code', 'dirty shortcuts', 'cutting corners'. The financial metaphor it's named after is actually more useful. Debt is a tool. Manage it like one. https://blog.turboawesome.win/2021/07/hiring-senior-engineers-what-the-interview-loop-cant-tell-you/ Wed, 07 Jul 2021 14:02:00 +0000 https://blog.turboawesome.win/2021/07/hiring-senior-engineers-what-the-interview-loop-cant-tell-you/ After running hiring loops for senior engineers at the European fintech firm, the pattern was clear: the things that determined whether a hire was successful were almost never well-measured by our interview process. https://blog.turboawesome.win/2021/04/go-1.18-generics-real-use-cases-worth-the-complexity/ Wed, 07 Apr 2021 10:29:00 +0000 https://blog.turboawesome.win/2021/04/go-1.18-generics-real-use-cases-worth-the-complexity/ Go 1.18 shipped generics and the ecosystem immediately split between 'finally' and 'we didn't need this.' After using them in production code, here are the cases where generics genuinely simplify things and the cases where they add complexity for no gain. https://blog.turboawesome.win/2021/02/writing-technical-rfcs-that-actually-get-read/ Wed, 24 Feb 2021 09:07:00 +0000 https://blog.turboawesome.win/2021/02/writing-technical-rfcs-that-actually-get-read/ Most technical RFCs fail at their primary purpose: creating alignment before building. The failures are almost always structural — missing context, buried decisions, or no clear ask. Here's the format that worked.