Service 01 / 05 · AI Engineering · all services
Stop paying per token for a task that has not changed: a small model on your own server
Distill, fine-tune and deploy the smallest model that does the job, then wire it into the place your team already works.
Most AI budgets go to frontier-model API bills for tasks a 300M-parameter student can do in a second on a CPU. The work is to find that student: build the corpus, distil or fine-tune, quantize, measure against the teacher, and ship the result behind an interface your team can operate. When a smaller model does not transfer, that is a published result too, and it saves the next month of your budget.
This is the right call when
- 01
Your frontier-API bill grows every month for a task that has not changed since spring.
- 02
Your team already lives in Slack or Claude and wants the CRM, the CMS and the reports reachable from there.
- 03
You need an agent that acts on company data, with every write audited.
What you get
What lands on your account
- Eval suite on your data, with the negatives published
- The smallest model that passes it, quantized, deployed on CPU or your account
- MCP server exposing your systems as tools, one write-gated gateway
- Handoff pack: how to retrain, re-evaluate and roll back
How it runs
Three steps. The six-step version, with the brief and the video request, is on the home page.
- 01
Scoping call
30 min, freeYou describe the situation; I ask three questions and propose a shape.
- 02
Savings audit
€1,200 / $1,400 net, 5 business daysThe 36-month cost comparison, the order of moves, the risks and a fixed build price. Paid by bank transfer after the call, once you accept the offer; credited in full against a build ordered within 60 days of the report. If the audit does not show savings of at least twice its fee over 36 months, you get the fee back.
- 03
Build, handoff, operate
fixed price from the auditOne engineer builds every layer on your account and documents it. The offer includes a monthly plan by default, so the system is looked after from launch day.
Case studies
The systems behind this service, with the numbers their owners cleared for publication.
H2M ML LabThirteen distillation experiments on real corpora. Students from 230M to 1.2B parameters, quantized, about a second on CPU.
Published negatives included: a distillation that did not transfer, and a teacher that turned out to be the cap.
- 13
- experiments
- 230M-1.2B
- student sizes
- ~1s
- CPU inference
AdtripSlack-native AI operator for a 25-person agency: headless Claude Code behind one write-gated gateway.
Every write to the CRM goes through one gateway with an audit trail; the brand voice was distilled from real sales calls, not a style guide.
- 2,315
- call transcripts distilled into brand voice
- 1
- write-gated MCP gateway
AdcareA clinic marketing operating system operated from Claude through its own MCP server.
Multi-tenant CMS plus agency and per-clinic CRM, every action available as a tool.
- 1
- MCP server over CMS + CRM
Runs on
Claude
Anthropic
MCP
Hugging Face
Pytorch
Python
Workers
Typescript
Articles
Read the math first
The numbers behind this, written down with their sources.
Make your business system operable by AI: lessons from hundreds of MCP tools in production
What an MCP server does for a business, how to gate write tools, why the audit log matters and what it costs. Lessons from 397 tools in production.
Questions I get about this
Tap a question to see the answer.
Scoping calls · most weeks
Whatever passes the eval on your data. Claude for agents and code; for narrow tasks often a small open model fine-tuned on your corpus and run on CPU. The comparison is published with the result.
Training corpora and fine-tuned weights stay in your storage. When a frontier API is used, it is under your own key with a written data-retention setting.
Savings audit. In 5 business days: the 36-month cost comparison, the order of moves, the risks and a fixed build price in writing. It costs €1,200 / $1,400 net, paid by bank transfer against an invoice after the call and after you accept the offer. If you order the build within 60 days of the report, the full amount comes off its first instalment; if not, the report is yours to take to anyone. If the audit does not show savings of at least twice its fee over 36 months, you get the fee back.
The repository, the Cloudflare account, the domains and the data sit under your company from day one, and the data is yours. Code built only for you becomes yours once it is paid for, as the contract sets out; H2M modules come with a perpetual licence. I get access to your accounts, not ownership of them.
Something else? Ask it on the call
A 30-minute call, free
Thirty minutes, then an offer for the audit
Pick a free time in my calendar. Or leave an email: I reply within one business hour with three questions and a proposed time for the call, not a deck.