What I bring to Group Lead and above

Management, quality and the software development life cycle come first, then AI, architecture and delivery. Each area says what I do, how I do it, and one example.

The software development life cycle (SDLC)

What I do: I own how work moves from idea to production: planning and slicing, code standards, code review, CI/CD and quality gates, releases, incidents and post-mortems.
How: Design docs and architecture reviews before the code. Written standards for scalability and backward compatibility. Delivery tracking so that slippage shows up early, while it can still be fixed.
Example: At an industrial IoT scale-up I set up architecture reviews, design docs and delivery tracking. Architecture issues started being caught in review instead of in production.

Management

What I do: Hiring, 1:1s, performance reviews, and growing senior engineers.
How: Weekly 1:1s. Formal performance reviews every six months. A structured hiring process.
Example: I have hired 15+ engineers and led teams of 3 to 8 across four companies. Several engineers I mentored were promoted.

Quality

What I do: I build quality into how the team works, not as a step at the end.
How: Code standards, code review, and CI/CD quality gates on every change. Design docs and architecture reviews before the code. Written standards for backward compatibility.
Example: I set up architecture reviews, and architecture issues started being caught in review instead of in production. I also built a breaking-change detector that runs on every pull request.

AI inside the team

What I do: I bring AI into the way a team builds, with rules, not as a free-for-all.
How: Claude Code for development, AI-assisted code review, and automated checks on every pull request. I have also built LLM document pipelines on AWS Bedrock: extraction, classification, structuring.
Example: I rolled out Claude Code and AI code review in my current squad, and built a breaking-change detector that runs on every pull request.

AI agents in production

What I do: I build agents that do real work, not demos.
How: An agent gets a narrow job, typed tools, and a boundary it cannot cross. Its output is checked by code, not by hope.
Example: A personal project of mine runs an agent that creates entities, answers analytics questions and handles support. A daily agent feeds the private side of this site through an authenticated API with scoped keys.

Architecture and scale

What I do: Event-driven systems, IoT and MQTT, AWS and Kubernetes, several data stores in one system.
How: Fix the constraint that blocks scaling first. Usually that means idempotency, ownership of state and clear contracts between services.
Example: I drove a move to an event-driven architecture and solved the idempotency issues behind it. Services that could only run as a single production instance could then auto-scale.

Delivery and stakeholders

What I do: Working with Product on priorities, and making trade-offs between technical debt and features explicit.
How: I put the trade-off on the table with its cost, then let the priority be a shared decision.
Example: At a fintech unicorn my team moved to the program launching paid subscriptions. We owned the frontend and I coordinated the integration across the payment, subscription and app backend teams.

Ready for Group Lead

What I do: I have not managed managers yet. I say so up front.
How: What prepares me is the work around that role: running several streams in parallel, leading cross-team projects, coordinating across teams and sites, setting the standards for technical reviews, standardizing architecture documents, taking part in other teams' architecture discussions, mentoring people into bigger roles, and building tools that serve more than one team.
Example: Coordinating priorities with Product, architects and engineering teams abroad on a global cloud integration. Several mentored engineers promoted. A pull-request check that works for the whole team, not for one person.

Match your job Ask my AI