LLM, agents and RAG consulting for companies

I am Riccardo Gasparini, a freelance AI engineer in Modena, Italy. I bring language models into company processes: search over internal documents, data extraction from documents and forms, agents that query existing databases and systems. With cloud models or self-hosted on GPU, when data cannot leave the company.

Typical problems

  • Thousands of technical documents, procedures and manuals that no one can query.
  • Data locked in PDFs, scans and Office files that must be mapped to a schema without errors.
  • A legacy database that only two people know how to query in SQL.
  • Models that cannot see sensitive data: on-premise inference is required.
  • A RAG prototype that "works in the demo" but no one has measured.

What I deliver

  • An ingestion pipeline with document parsing (Docling, OCR, vision-language models) and semantic chunking.
  • Hybrid dense and sparse retrieval with reranking, visual page retrieval when needed.
  • LangGraph agents with structured, validated output (Pydantic), text-to-SQL with guardrails.
  • Evaluation with RAGAS or proprietary test sets, before and after every change.
  • A multi-tenant FastAPI backend with authentication, streaming, Docker or Kubernetes deployment.
  • Self-hosted serving with vLLM or Ollama, or the Claude and OpenAI APIs.

How I approach it

I start from the documents and the users' real questions, not from the model. I build an evaluation set within a few days, then iterate while measuring. In production every component is replaceable: model, vector store, reranker.

Email meLinkedIn

Work in this area

Industrial group · 2026

Multi-company RAG platform over internal documents

Problem
Thousands of technical documents and procedures scattered across the group's companies, impossible to query, and no data allowed to leave the company.
What I did
Ingestion with Docling and semantic chunking, triple indexing (chunks, generated questions, summaries), hybrid dense and sparse search with cross-encoder reranking and visual page retrieval with ColPali. Fully local LLM stack on vLLM, streaming answers with per-company permissions, continuous evaluation with RAGAS, Kubernetes deployment.
  • Docling
  • Qdrant
  • BGE-M3
  • ColPali
  • vLLM
  • RAGAS
  • FastAPI
  • Helm

Laboratory diagnostics · 2026

Structured extraction from lab reports

Problem
Parameters like sodium and potassium, with values and units, locked in PDF and scanned reports with labels that vary every time, to be extracted without making anything up.
What I did
A three-agent LangGraph graph: one normalizes the requested terms, one searches the text for the parameters with exact or fuzzy matching, one validates and corrects the final JSON with found and not-found lists. Pydantic output, streaming, per-user document archive, self-hosted models, multi-tenant on Kubernetes.
  • LangGraph
  • Pydantic
  • Docling
  • vLLM
  • FastAPI
  • Kubernetes

Institutional catering · 2026

Conversational assistant over a legacy recipe database

Problem
An IBM DB2 database of recipes and ingredients that only SQL writers could query, with staff asking questions in natural language.
What I did
Three LangGraph agents: the first classifies intent, the second generates safe DB2 queries (dynamic WHERE clauses, case-insensitive search, joins optimized for version 9.7), the third filters and aggregates with pandas. Expiring sessions, KPIs, CSV export, an open model on self-hosted vLLM, Keycloak authentication, Helm and CI.
  • LangGraph
  • IBM DB2
  • vLLM
  • Keycloak
  • FastAPI
  • Helm

Frequently asked questions

What kind of projects do you work on?

Industrial computer vision projects, LLM-based systems (RAG, agents, structured extraction), forecasting on IoT and geospatial data, and web and mobile platforms that turn these models into a product. From PoC to production, with code, tests and deployment.

Do you use cloud or on-premise models?

It depends on the data. When it can leave the company I use APIs like Claude or OpenAI; when it cannot, I run open models on your GPUs or on a dedicated server with vLLM or Ollama. I have put both approaches into production.

How do we get started?

Send me a couple of lines about the problem. We do a free thirty-minute call to see if it makes sense to work together. If so, I put together a proposal with the goal, success metric, timeline and cost.