Skip to content
services

What you need,
built for production.

From architecture to production code: three ways to solve a technical problem without bloating the scope.

three ways to work together

For each: who it’s for, what the engagement involves, and what you get at the end.

01

AI & automation

who it’s for

Teams and companies with well-defined processes that are ready to be automated.

Systems built with LLMs and AI agents that take repetitive work off your team’s plate.

01Process analysis: what to automate and where human judgment still matters
02AI pipelines designed and shipped to production
03Documentation and handoff to your internal team
02

Software development & architecture

who it’s for

Teams moving from a prototype to a system that can support real users.

Backend, frontend, and real-time systems for products built to hold up over time.

01Architecture designed to evolve with the product, reducing the risk of repeated rewrites
02Tests and CI shipped with the code, so your team can change the system with confidence
03No black boxes: code, infrastructure, and technical decisions documented so your team retains control
03

Technical advisory

who it’s for

Founders and CTOs who want an outside perspective before committing to a technical direction.

A direct review of your stack, architecture, and roadmap. I tell you what I would keep, rework, or drop, with the trade-offs made explicit.

01A clear view of what you have: what is solid, what is fragile, and what can wait
02Concrete options, each with its cost and rationale
03A prioritized list ranked by cost and return, ready to bring to the board
the stackShipping AI systems since GPT-3, before ChatGPT

The tools of the trade.

backend
Primarily Python

Django and real-time data pipelines, with the rest of the stack chosen to fit the product rather than forced into a single framework.

frontend
TypeScript, React, Next.js

Clean Tailwind CSS interfaces, from UI implementation through deployment.

ai
LLMs and multi-agent systems

Document extraction, embedding and reranking pipelines, and self-hosted open-source models on NVIDIA GPUs. Self-hosting can keep data and infrastructure under your control and make costs easier to forecast.

infrastructure
Docker and Terraform

Docker and Docker Compose on every project, with Terraform for versioned, reproducible infrastructure.

how I workarchitecture firstreadable codea clear rationale for every choice
process

How we get to production.

01
Discovery and analysis

I start with the context: goals, constraints, and existing code. I ask questions until the problem is clear.

02
Clear proposal

Scope, timeline, and price in writing. Engagements are either fixed-price projects or ongoing collaborations with an agreed monthly capacity. In both cases, deliverables, responsibilities, and acceptance criteria are clear before work begins.

03
Iterative delivery

Short iterations and working releases. You see progress every week, not just at the end.

04
Clean handoff

Documentation, tests, repository access, and a clean handoff to your team, so the system remains understandable and under your control.

when to call me

When to call me. And when not to.

It makes sense if
A manual process eats up hours of your team’s time every week.
You need to go from a prototype to a system that holds up under real-world use.
You want an outside technical opinion before committing to a direction.
You have an idea to bring to market and need someone to build it end-to-end.
It’s not a fit if
You need a brochure site by tomorrow.
You’re looking for a large team with account managers.
You need short-term staff augmentation rather than ownership of a defined outcome.
contact · the first step is a call

The first step is a call.

Tell me about the problem and we’ll see whether I can help. Whether you’re starting from scratch or already have a live system, the first step is the same.

Google Meet · 30 minutesinstant confirmationscope, timeline and price in writingdirect, no strings attached