Case study · Enterprise operations
Support assistant in Microsoft Teams
A Teams and telephony assistant that answers support questions from the company’s own documents, using an OCR and entity-extraction pipeline underneath.
- Client
- Customer-support and internal-operations teams
- Result
- Live in a corporate Teams tenant
- Stack
- LLM APIs, Microsoft Teams, Vision OCR, Named-entity recognition, Retrieval (RAG)
The problem
Support and operations staff kept answering the same questions out of long, unstructured documents. The assistant had to cite that material, so a general-purpose chatbot was ruled out.
The work
- A tool-using assistant deployed in Microsoft Teams, including telephony.
- A Vision OCR and document NER pipeline that extracts entities from unstructured documents.
- Retrieval over the extracted data, so each answer comes with its source.
Outcome
The assistant is live in a corporate Teams tenant and answers from the organization’s own documents.
The client isn’t named, to respect confidentiality. I can share more detail on a call where permitted.
Talk about a similar projectMore work
AI due diligence for investors
Technical reviews of early-stage AI startups for venture investors, plus retrieval tools that turn a data room into a first-draft investment memo.
Seed and Series A deals
From LLM wrappers to agents
Design reviews and workshops for venture-backed engineering teams moving from a single LLM call to agents built on LangGraph, CrewAI or AutoGen.
Several portfolio companies