Agentic & adaptive RAG
Certilab RAG for calibration certificates
A certification product and RAG foundation that turns calibration certificates into searchable, traceable decision support.
Role
Freelance Full Stack & AI Developer at Certilab
Stack
Laravel 11 · Filament · MySQL · S3 · Python · LangGraph · FastAPI · Qdrant
Problem
Calibration certificates combine structured fields and document content that users need to search, understand and validate.
My contribution
I designed agentic and adaptive RAG services with structured/semantic search, source citations, access control, query rewriting and answer verification.
Architecture & stack
LangGraph, FastAPI, Qdrant, OpenAI and MySQL; the adaptive version works over 3,848 indexed vectors from real certificates.
Result
A certification system with 300+ certificates issued, plus a production-oriented RAG and chatbot foundation.
From product foundation to AI layer
I built the certification foundation with backoffice workflows, customer access, QR validation, signed PDFs, roles and auditability, then extended it with agentic and adaptive RAG services for certificate questions.
Designed for traceability
The RAG projects combine structured and semantic retrieval, source citations, access control and answer verification. The adaptive public project documents a seven-node graph with query rewriting and hallucination-check loops over pre-indexed certificate vectors.
From certificate to decision support
01
Certificate system
Issuance, signed PDFs, QR validation, roles and audit trail.
02
RAG services
Structured and semantic retrieval with citations and guarded access.
03
Business questions
Chat and search experiences over certificate data and documents.