AI assistant for manufacturing
Technical specs, work instructions, quality procedures and maintenance manuals run to thousands of pages. People on the floor either search for a long time or ask someone.
▶ The solution
I build an AI assistant that works on your procedures, reports and past projects, answers with sources, and is mathematically optimized for speed and low cost.
▶ Use case
When a maintenance technician asks about a fault code, the assistant finds the relevant manual section and past maintenance notes and summarizes them with sources.
▶ Data you need
Technical documents, work instructions, quality procedures and maintenance records.
The industry's data landscape
In manufacturing, data usually lives in the ERP (orders, stock, purchasing), in production tracking (downtime, scrap, shifts) and in quality records. These systems rarely talk to each other; even the same product can carry different codes in each. So the first job is often to bring them together in one table.
An in-house AI assistant connects large language models (LLMs) to your company's own documents. An employee asks a question; the assistant finds the relevant documents, writes the answer based on them, and shows where it came from.
The math behind it
Monthly cost ≈ N × (t_in × p_in + t_out × p_out)
- N: questions per month
- t_in, t_out: tokens sent to and returned by the model per question
- p_in, p_out: the model's price per token
How I work
- Week 1
Use case and sources.
Who is the assistant for, and for which questions? Which documents are in scope, and which are current? How success will be measured is written down here.
- Weeks 2–3
First working version.
Documents are ingested, the assistant is set up and answers with sources. It is tested with real users' real questions.
- Weeks 4–5
Accuracy and trust.
Answers are measured regularly against a question set. Wrong or incomplete answers are fixed, and the assistant is made to say when it doesn't know.
- Final week
Handover.
How to add new documents, how to monitor the system and how to guide users is documented and handed to your team.
▶ Questions and answers
AI assistant for manufacturing — what data do you need?
Technical documents, work instructions, quality procedures and maintenance records.
When is this the wrong choice?
You have few documents and everyone knows where they are. A well-organized folder structure and search may be enough. An assistant would be an unnecessary layer. The knowledge is in people's heads, not in documents. No assistant can find knowledge that was never written down. It has to be written down first.
How does this work start?
With a short intro call. I listen to what you need and tell you plainly whether your data is enough for the job and what the first step should be. I reply to every message within 24 hours.
▶ Concepts
Linked to Wikipedia and Wikidata.
▶ Sources
- [1]Wikipedia, Retrieval-augmented generation, 2026-10. en.wikipedia.org/wiki/Retrieval-augmented_generation
- [2]Wikipedia, Large language model, 2026-10. en.wikipedia.org/wiki/Large_language_model
▶ Intro call
Let's talk about your situation.
In a 30-minute call I listen to what you need and tell you plainly how I can help.
Book an intro call▶ In other industries