I ship AI agents, RAG systems, skills, workflows and copilots in production.

AI agents, skills, RAG and automation wired into your existing systems, shipped to production, cloud or on-prem. Token spend is decided at design time.

Field
Tell me about your case

Working together

How many days, and what you have at the end.

Durations are in days, set in the quote alongside the price: these are commitments, not estimates. The right format is settled on your case, not on a catalogue. By default, it's the audit.

  • AI Audit

    Start here4days

    An honest lay of the land: where AI would pay off, where it would just burn money, and in what order to tackle it.

    Who it's for

    You're unsure where to start, or you have doubts about a project already underway.

    What you get

    • Your processes put under the microscope
    • Use cases prioritized by ROI
    • Pitfalls and costs flagged early

    Deliverable

    A clear report and a costed roadmap.

  • POC Sprint

    12days

    A prototype that proves the value on your real data, with success criteria set up front.

    Who it's for

    You want to de-risk before investing in a full build.

    What you get

    • One targeted use case
    • A prototype on your real data
    • An evidence-backed go/no-go decision

    Deliverable

    A working POC and the numbers to make the call.

  • Production Build

    20+days

    A reliable, monitored, documented system in production that your teams keep using.

    Who it's for

    The use case is proven; now you need a system that runs for real.

    What you get

    • A deployed system, cloud or on-prem
    • Evals, guardrails, observability
    • Documentation and handover

    Deliverable

    A production-ready product that you own outright.

  • Embedded AI architect

    10+days · renewable

    The AI expertise your team is missing, on an agreed volume with an end date set in the contract.

    Who it's for

    You already have a team and a project underway, but nobody to settle the AI architecture.

    What you get

    • Architecture decisions settled with your team
    • Code shipped, not recommendations
    • Your developers skilled up

    Deliverable

    Code in production and a documented architecture.

What I carry

Written into the proposal, not promised out loud.

A first client handing a budget to an independent he doesn't know is taking a risk. Here is the share of it I take on, and what you can hold me to.

Tell me about your case

First conversation is free, no strings attached. At the very least you leave with a straight answer and a ballpark on your case.

Who it's for

For you, or not at all.

I would rather waste thirty seconds of your time here than three weeks of it in scoping. Here is who I genuinely help, and who I don't.

For you if

  • You have a technical team or an IT department, and data that is yours.
  • An AI prototype works in a demo but never reaches production, or doesn't survive there.
  • You need to wire AI into existing systems without opening an access hole.
  • You have sovereignty or confidentiality constraints that rule out all-cloud.
  • You want to keep the system and evolve it yourself after delivery.

Not for you if

  • You are looking for a white-label chatbot to drop onto a brochure site.
  • You want a long-running staffing arrangement with no defined objective or end date.
  • Your data doesn't exist yet, or nobody knows where it is.
  • The deciding criterion is lowest price: I won't be the cheapest.

If you are somewhere in between, the audit settles it in four days and commits you to nothing beyond it. And if your case isn't a fit, I tell you then, before any invoice.

Work

AI systems in production.

A few recent projects: MCP infrastructure, an assistant under human sign-off, anomaly detection, RAG. The starting problem, what I built, and what it changes.

    • AI infrastructure
    • Security

    Give AI agents access to internal tools without opening a hole.

    Every LLM wired to each internal tool by hand

    Teams wire an AI onto an internal tool without opening extra access

    MCPSkillsfastmcpZero TrustSSO
    • Agents
    • Human sign-off

    Have an agent write the architecture record it cannot confirm.

    Records retyped by hand from one deliverable to the next

    An entry nobody has confirmed reaches no report and no client link

    LangGraphFastAPIPostgreSQLReactTOGAF
    • Anomaly detection
    • Cost

    Price the capacity paid for nothing, with no model inventing a number.

    Capacity thresholds watched by hand, and nothing priced

    The capacity gap is priced by month and year, with the sum shown

    PythonFastAPIscikit-learnNext.jsLLM
See the work

Work done on assignment. Full context on request.

Why guinat

Why trust me with your AI.

Nathan Guihot

Nathan Guihot

AI and agentic systems specialist

[email protected]

  • Shipped and adopted

    I ship systems that run for real and that your teams keep using, not POCs that gather dust.

  • ROI first

    I only do AI where it pays off. I'll advise against the cases that aren't worth it.

  • Your data stays yours

    Cloud or on-prem, EU hosting, private models: your data stays inside your walls.

  • One point of contact

    I'm Nathan. From audit to deployment, you talk to the person who designs, codes and ships, not a middleman.

  • Costs under control

    Tokens are expensive. Caching and the right model in the right place: 30 to 70% savings is the range I target, measured on your own bill.

  • Reliable and measured

    Evals, guardrails, observability: systems you can trust, not black boxes.

Let's talk

Let's talk about your use case.

Tell me about your case[email protected]

Free, no-obligation first exchange. At the very least you walk away with a straight opinion.