Why I write

After two decades shipping enterprise platforms and consumer experiences, I realized the conversations that move projects forward rarely make it into public view. The retrospectives, trade-off diagrams, and hard-won fixes stay buried in notebooks. This site keeps that door open—so leaders, engineers, and curious builders can reuse the playbooks instead of starting from zero.

AI Engineering Playbooks

A major focus of my recent work is practical AI engineering. I build playbooks, runnable demos, and open-source repos that make AI systems easier to understand, evaluate, and apply in real delivery environments. The goal is not to publish vague AI thought pieces, but to show the actual mechanics behind generation, retrieval, orchestration, prediction, and evaluation.

That portfolio currently includes:

  • Practical AI Engineering Playbooks as an umbrella view across the patterns teams most often confuse in delivery settings.
  • Generative AI Engineering Playbook for structured generation, prompt contracts, and mock-first workflow design. GitHub repo.
  • RAG Engineering Playbook for grounded retrieval, citations, threshold checks, and anti-hallucination thinking. GitHub repo.
  • Agentic AI Engineering Playbook for multi-step orchestration, context handoff, and role-based workflow execution. GitHub repo.
  • Model Engineering Playbook for predictive workflows that train, evaluate, save, and serve models from labeled examples. GitHub repo.
  • AI Quality Engineering Playbook for evaluation discipline, test generation, retrieval-aware quality workflows, and release-confidence guardrails. GitHub repo.

Together, these playbooks reflect the direction I care most about right now: practical AI systems that can be reasoned about, tested, and reused by real engineering teams.

How I work

During the day I lead product engineering initiatives: shaping strategy, aligning UX outcomes, guiding architecture decisions, and helping teams ship with confidence. Outside that core delivery work, I invest heavily in reusable engineering assets—open-source tooling, AI playbooks, and automation systems that turn implementation lessons into artifacts other teams can actually adopt.

That work spans several layers:

  • AI engineering: practical playbooks and demos for generation, retrieval, orchestration, model workflows, and quality evaluation.
  • Quality engineering: WebdriverIO, Playwright, and reusable test generation patterns that reduce friction in release workflows, including wdio-testgen-from-gherkin-js, wdio-testgen-from-gherkin-ts, playwright-testgen-from-gherkin-js, and playwright-testgen-from-gherkin-ts.
  • Platform and API tooling: projects like Restlyn that support API-first workflows and reusable automation.
  • Open-source accelerators: Gherkin-based generators for WebdriverIO and Playwright, plus earlier integration toolkits that solved real delivery gaps.

Across all of this work, the through-line is consistent: build dependable systems, make delivery knowledge reusable, and turn hard-won implementation lessons into practical engineering playbooks. For a broader view of how those ideas move from concept to working software, explore my GitHub repositories.

What you can expect here

Each article is grounded in real delivery work. You'll find:

  • AI engineering playbooks that break down practical patterns across Generative AI, RAG, Agentic AI, Model Engineering, and AI Quality Engineering.
  • Quality engineering playbooks that keep releases calm—automation patterns, observability, evaluation, and reliability rituals.
  • Software development journals that track design spikes, architectural decisions, and the craft behind production launches.
  • Product & engineering leadership notes on shaping roadmaps and empowering teams.
  • DevOps and SRE field guides that translate platform runbooks into repeatable playbooks for your teams.
  • Experiments across AI/ML, cloud platforms, and tooling that make software craftsmanship more humane, testable, and scalable.
  • Occasional deep dives into UX practices that ensure users feel the improvements we ship.

Dig in next

If you're exploring for the first time, these collections are a good starting point:

Let's connect

I'm always open to pairing on an experiment, speaking with teams, or swapping stories about the craft. Send a note and let's keep the conversation going.