Skip to main content
Dimension 02 // AI & automation

AI with a job description, not just a demo.

We build AI systems that do specific, measurable work: read invoices, answer questions from your documents, forecast demand, and flag what needs a human. Then we wire them into the tools your team already uses.

Treat the model as one unreliable component in a reliable system. Measure everything.

Patterns
Extraction · RAG · Agents
Deployment
Cloud APIs or private models
Safety
Human review built in

Explore ai & automation

The gap between demo and production

Every AI demo works on the presenter's laptop. Production is where documents arrive sideways, users ask questions nobody anticipated, rate limits kick in at month-end, and a model update quietly changes behavior.

So we design AI features like any other critical system: structured outputs, validation rules, confidence thresholds, human review for uncertain cases, logging of every call, and evaluation sets that catch regressions before users do.

Principles we don't skip

Structured outputs

Schemas and function calling, never parsing free-form prose in production.

Observability

Every prompt, response, latency, and cost logged and traceable.

Humans in the loop

Low-confidence results route to reviewers instead of guessing.

Adversarial testing

Red-team datasets built before launch, not after the first incident.

Local synthetic preview // No file upload or OCR

Explore an extraction review

Run an invented sample through a local preview endpoint to explore the review flow. It does not accept documents, run OCR, or send data to an external service.

Document preview

Synthetic sample

Sample invoice

Run the preview to see sample details

Illustrative text only. No real business or person is represented.

Illustrative extracted fields

Preview not run
FieldSample valueSignal

    Run the preview to load illustrative fields.

    The signal values are invented endpoint fixtures, not model measurements. A real workflow needs an OCR service, validation rules, and a human review policy.

    How we approach AI projects

    We treat a model as one component in a larger system. It reads, classifies, drafts, or predicts, and everything around it is ordinary, reliable engineering: validation, permissions, logging, review queues, and integration with the tools people already use.

    Every project starts with a specific job and a way to measure it. We build an evaluation set from your real data before choosing a model, so decisions about models, prompts, and thresholds are based on numbers rather than impressions. When a simpler rule or a standard integration solves the problem better, we recommend that instead.

    Which service fits your problem

    If you want toStart with
    Stop retyping invoices and formsDocument intelligence
    Let staff ask questions of company documentsRAG knowledge systems
    Automate multi-step back-office tasksLLM agents
    Add suggestions inside Odoo screensAI in your ERP
    Buy the right stock at the right timeDemand forecasting
    Find the use cases worth doing firstAI readiness assessment

    Principles behind every AI system we build

    • A clear job with a measurable outcome
    • An evaluation set from your real data
    • Humans review uncertain or risky results
    • Least-privilege access for every component
    • Full logging of inputs, outputs, and actions
    • Costs measured per task and monitored

    How an AI engagement runs

    1. Discover

      Find the job, the data, and the measure of success.

    2. Evaluate

      Build an evaluation set from real examples before choosing a model.

    3. Pilot

      Ship a supervised version to a small group and measure it.

    4. Scale

      Increase automation where the numbers support it, and monitor in production.

    Where AI usually pays off first

    The best first AI projects are boring in a good way: high volume, repetitive, and easy to check. Reading supplier invoices, sorting incoming requests, answering questions from internal documents, and drafting routine replies all fit that pattern.

    They share one property: a person can quickly confirm whether the AI got it right. That makes it possible to measure quality from day one and to widen automation only where the numbers support it.

    Your data, your rules

    We design AI systems so your data stays under your control. Depending on the use case, that can mean providers configured not to train on your data, models hosted in your own cloud account, or open models running on your infrastructure. The choice is documented, with the trade-offs in cost and quality spelled out.

    Questions to ask any AI vendor

    • How will we measure accuracy on our own documents or questions?
    • What happens when the model isn't sure?
    • Where is our data processed and stored, and is it used for training?
    • Who can see what the system did, and why?
    • What does it cost per task at our real volumes?

    Not sure where to start?

    The AI readiness assessment ranks your options by value and feasibility, and names one pilot worth doing first.

    Frequently asked questions

    Which AI models do you use?
    Whichever fits the task, cost, and data constraints: hosted frontier models, smaller open models, or classic machine learning when that's simpler.
    Can AI run on our own infrastructure?
    Yes. Private deployments with open-weight models are an option when data must not leave your environment.
    How do you measure whether AI is working?
    With evaluation datasets and business metrics agreed up front: accuracy per field, review rates, time saved, and error costs.
    Will AI replace our team?
    Our systems remove repetitive work and route judgment calls to people. The goal is fewer copy-paste hours, not fewer colleagues.
    How long does a first AI project take?
    A focused pilot on one workflow typically starts with a short discovery and a prototype evaluated on your real data.
    Which AI models do you work with?
    Hosted models from the major providers and open-weight models we can run privately. We choose per task, data policy, and cost.
    What does an AI pilot usually involve?
    One narrow, high-volume task, an evaluation set built from your real examples, a review step for people, and clear success measures agreed before we start.
    Can AI connect to Odoo?
    Yes. Common examples include reading supplier invoices into Odoo, answering questions from Odoo data with the user's own access rights, and forecasting demand from sales history.