May 28Member-onlyUnfocused large-scale transformation is bad strategy.I’ve previously said that I prefer simultaneous coordinated and opportunistic change strategies. The opportunistic change strategy was about enabling mass power; the coordinated change strategy was about focusing and directing power. I also said that if I couldn’t do both, I would choose the coordinated, focused strategy. Let’s expand on…Change Management2 min readChange Management2 min read
May 21Member-onlyKey practice: Evolutionary designKey practice in effective product development culture. — TL;DR Evolutionary design is YAGNI + Simple Design + Refactoring. Also known as Emergent design; Continuous design; JIT (just-in-time) design Opposites Big Design Up Front; Planned design; Anticipatory design YAGNI (You Ain’t Gonna Need It) YAGNI means building features only when they are necessary. …Evolutionary Design2 min readEvolutionary Design2 min read
May 21Member-onlyWhich is more motivating: Relatedness or Purpose?Migrated from my old blog. — When talking about motivation, many people will reference Drive with its three key factors: Autonomy Mastery Purpose One of the main sources that Dan Pink used for Drive is self-determination theory, which also points to three key factors for intrinsic motivation: Autonomy Competence (which…Motivation1 min readMotivation1 min read
May 6Member-onlyKey practice: Test Driven DevelopmentKey practice in effective product development culture. — Test Driven Development: Write a single test first; Run the test to ensure it fails; Write just enough code to make the test pass; Refactor (see Kent Beck’s 4 rules of simple design) ; Repeat This is often summarised as Red Green…2 min read2 min read
May 3Member-onlyAssessing change models using 6 sources of influenceI’ve previously written about the 6 sources of influence: Let’s use this to assess some well-known change models. Kotter 8 step change model From The 8-Step Process for Leading Change | Dr. John Kotter (kotterinc.com) but also Leading Change, The Heart of Change, and The Heart of Change Field Guide and a whole bunch of…Change2 min readChange2 min read
Published inGoPenAI·Apr 29Collecting useful Large Language Model (LLM) referencesI’m learning more about LLMs and I figured I’d record useful references as I go. I’d appreciate any additional suggestions. At the moment, I’m approaching this using 4 categories: Prompt engineering; Fine-tuning LLMs; How LLMs work; Broader topics. Prompt engineering ChatGPT Prompt Engineering for Developers by DeepLearning.AI which seems to be the…ChatGPT3 min readChatGPT3 min read
Apr 27Member-onlyKey practice: Aligned, autonomous cross-disciplinary teams.Key practice in effective product development culture. — Effective product development teams have clear missions that are aligned with a broader overall product strategy. They are expected to act autonomously and empowered with the capability and support to do so. This typically requires the teams to be cross-disciplinary. Autonomy is about enabling teams to decide and act more quickly. Teams that…Team4 min readTeam4 min read
Mar 25Member-only4 simple rules for team designI’ve been wondering about simple rules for team design inspired by Kent Beck’s rules for simple design. This is partly triggered by noticing people incorrectly using team size as their primary (or even sole) design criteria. — In order of priority, a well-designed team: Has a clear mission. Everyone knows…Team1 min readTeam1 min read
Mar 6Summary of my job search (Dec 2022 — Feb 2023)On 29 Nov 2022, I learned I was being laid off by Spotify, official termination date 31 Dec 2022. This is a rough summary of my job search experience. I started looking the moment I got the same day “urgent” calendar invite on 29 Nov and had set my LinkedIn…Job Search2 min readJob Search2 min read
Feb 27Member-onlySome thoughts on reducing the likelihood of failure with technical programs.Use an iterative-incremental lifecycle. Using a serial lifecycle (aka waterfall or SDLC) increases the likelihood of failure. If you are using a serial or waterfall lifecycle, you are increasing the likelihood of failure. But what if the problem is trivial? Once you’re in “program” territory, it’s highly unlikely that the effort is trivial enough to risk waterfall. Note: Using a “hybrid waterfall” is still unnecessarily risking failure.Technical Program Manager4 min readTechnical Program Manager4 min read