01When it makes sense
- Situation
- Health, legal, financial data or trade secrets
- Provider API
- Possible with the right contracts, but data leaves
- Private model
- Data never leaves
- Situation
- Few requests per day
- Provider API
- Cheaper
- Private model
- Fixed hardware cost
- Situation
- High continuous volume
- Provider API
- Cost grows with usage
- Private model
- Predictable cost
- Situation
- You need the most capable model available
- Provider API
- Better
- Private model
- A step behind on the hardest tasks
- Situation
- Air-gapped environments
- Provider API
- Not usable
- Private model
- Works offline
02How the work runs
- Discovery. Before any code: interviews, a look at the data you already have, technical and business constraints on the table.
- Architecture. Data schema, component boundaries and failure points decided on paper, when changing them costs an hour and not a month.
- Build. Short iterations with working demos instead of one delivery at the end. AI speeds things up where it makes sense; it never replaces review and testing.
- Handoff & scaling. Documentation, runbooks and a system your team can run without depending on me. No lock-in.
03Pricing
Every project is priced after a scoping call: the same word, "chatbot" or "agent", can mean a two-week prototype or a system in production for thousands of users. What moves the price:
- model size and concurrent users, which set the hardware;
- on-site servers or EU cloud;
- integration with existing systems and permissions;
- support level after release.
The first call is free and ends with a clear answer: whether the project makes sense, roughly how big it is, and what to do first.