Prototype · 2024
BOM Scriber
Recognising and normalising bills of materials (BOM) from files sent by customers.
- AI / LLM
- Business apps
- Documents and files
- Classification
- Data extraction
- Rules and validation
A system that receives a customer's file, decides whether it is a bill of materials and brings it to one format.
Problem
Every customer sends bills of materials in a different layout, and matching columns by hand takes time and causes quoting errors.
Solution
A language model recognises the sheet structure and matches the columns, and rules check and tidy up the result.
What the system does
- Recognising whether a file is a bill of materials
- Matching columns to one format: part number, quantity, manufacturer
- Validating and normalising the result
- Documents and files
- Classification
- Data extraction
- Rules and validation
Related projects
- Messaging channels
- Language model
- Knowledge baseHand-off to a person
- CRM
2024 · In production
Omnichannel AI Support
A SaaS platform with AI agents that serve a company's customers across several messengers at once.
- AI / LLM
- Backend
- Integrations
- Frontend
2026 · Pilot
ColdCallAI
A voice agent calls leads and a model extracts facts from the call — the score is computed by code, not AI.
- AI / LLM
- Backend
- Integrations
2026 · MVP
Olimpio
Competitions and olympiads gathered from the web and email, classified by AI and matched to each student.
- AI / LLM
- Backend
- Frontend
- Data & analytics