Help with the hard technical parts.
I help early-stage teams with the first platform, scaling it when traffic shows up, and running AI systems in production. I've done all three as a CTO and founding engineer, and I still write the code.
Tell me what you're buildingWays to work together
Pick one strategy: :one_for_one
-
restart: :transientAdvisory
A few hours a month on architecture, hiring, and the calls you'd like a second opinion on. Good for seed-stage teams.
back when you need it
-
restart: :temporaryFixed-scope project
An architecture review, a migration, or a performance problem, with a clear end and a written handover.
exits after the handover
-
restart: :permanentFractional CTO
Technical direction and your first engineering hires, until you're ready for a full-time CTO.
runs until you hire a CTO
What I work on
-
Your first platform
Picking the stack, setting up deploys, and writing a codebase your next engineers can work in. Usually Elixir and Phoenix, with Rust or Go where they fit.
-
Distributed systems
Event pipelines with backpressure, clustered Elixir, and Postgres or ClickHouse that keep up as load grows.
-
AI infrastructure
Whether to self-host at all, GPU sizing, inference tuning, and evals that catch a model quietly getting worse.
-
SRE and Nix
Kubernetes when you need it, Terraform, observability, and reproducible builds with Nix and devenv.
How it starts
-
both of us
A short call
We figure out if I'm the right person for it.
-
me
A written scope
What I'll do, for how long, and what you get at the end.