All posts
AI EngineeringAutomation

Odoo + AI: automating quote-to-cash without breaking finance

Where machine learning genuinely helps inside an ERP, where it absolutely should not go, and how to keep an auditable trail through both.

Deep Tensors TeamMay 22, 20261 min read

ERP automation projects fail in a specific way: someone points a model at a process finance depends on, and the first unexplained number ends the project.

Automate the reading, not the deciding

Extracting line items from a supplier PDF is a great use of a model. Deciding whether to pay it is not. Draw that line explicitly and put a person on the other side of it.

Everything writes an audit record

Every automated field carries who or what set it, with what confidence, from which source document. Finance will ask. The answer has to be one click away.

Confidence thresholds route work

High confidence posts straight through. Middling confidence lands in a review queue. Low confidence never touches the record. Tune the bands against how expensive a mistake actually is.

Measure hours, not accuracy

The number the business cares about is time removed from the process. Track it from day one, because it is what pays for the next phase.

Deep Tensors Team

Engineering

Notes from the team shipping AI, web, and automation systems for companies that need them to actually work on Monday morning.

Building something like this?

We help teams take AI systems from prototype to production — evaluation, infrastructure, and the interface around it.

Talk to us

Keep reading

AI EngineeringLLMs2 min

Why most AI agents fail in production

The demo works, the pilot stalls, the rollout quietly dies. Four failure modes we see in almost every agent project — and what a system that survives contact with real users looks like.

Ishtiyaq KhanJul 28, 2026
AI EngineeringLLMs1 min

RAG is not a product: retrieval patterns that hold up

Chunk, embed, search, stuff into a prompt. That pipeline gets you a convincing prototype and a support queue full of confidently wrong answers. Here is what we build instead.

Ishtiyaq KhanJul 14, 2026
AI EngineeringComputer Vision1 min

Computer vision on the factory floor: six lessons

Lighting changes, cameras drift, and the operators will move the mount. What we learned deploying defect detection into a plant that runs three shifts a day.

Deep Tensors TeamJun 30, 2026