AI prototyping & MVPs
Three-week sprints that produce a deployed system, an honest evaluation, and a go-or-kill decision with a cost model. Three founded companies and ten internal MVPs of practice at this exact shape.
- You have sat on an AI idea for months and need it shipped
- You want a deployed system in your stack, not slides
- You can give three weeks of stakeholder access and a real metric to chase
- You want a maintained product on day one — start with a Build engagement
- You do not know what to build yet — start with an Assessment
How this ships
One use case, one metric to move, one stack. Scoped in the first week, signed off before any code lands.
Deployed in your environment behind your auth. Evaluation harness and cost / latency profile attached, not bolted on later.
A defensible decision with the numbers to defend it. If the answer is build, the Build quote is ready the same day.
The work that backs this
The cases below are the ones I would point a sceptical buyer at first. The full bench lives on /work.
In-house LLM pipeline, $337K cheaper
Replaced an annual vendor translation contract with an in-house LLM pipeline. Shipped in 6 weeks. 12,000+ key-value pairs translated since Jan 2024.
Autonomous drone delivery in production
Founding engineer. Shipped the end-to-end software & cloud stack that powered 500+ last-mile drone deliveries. Partnerships with Unilever, showcase at CES Las Vegas.
Templating engine powering 15% of the platform
Designed and owned the distributed templating engine. Cut customer setup time by 90% and became the system 15% of all content on the platform is created from.
How I price it
Most engagements start with the first option. The second is the natural next step or a smaller starting bet.
Want to talk about ai prototyping & mvps?
Book a 30-minute call. I'll tell you straight whether this is the right shape for your problem.