Extract
PDF CV + job description → normalized candidate and role text
Evidence-Based AI · Full-Stack Application / v1.6.1
A full-stack AI application that compares a PDF CV with a job description, extracts role requirements, verifies candidate evidence, calculates a deterministic match score, and generates guarded application material with a local LLM.
PDF CV content and a complete job description are processed into a structured requirement-evidence table.
Requirements are classified into MATCHED, PARTIAL, and MISSING states before scoring.
A deterministic Python scoring layer produces a transparent 100-point breakdown.
The local LLM is kept out of the scoring path and used for guarded natural-language generation.
PDF CV + job description → normalized candidate and role text
Requirements → canonical evidence states
Deterministic Python weighting → explainable score
Structured evidence → guarded local application material
This case study is based on the project’s own documentation.
Agentic AI · Production Engineering
A production-oriented agentic workflow platform with structured decisions, tool calling, human approval, asynchronous processing, and AWS Infrastructure as Code.
Local Speech AI · Full-Stack Application
A privacy-first meeting assistant that transcribes recordings locally, extracts structured notes, and answers grounded questions.
Local RAG · Full-Stack AI Application
A privacy-first document assistant with local AI, streaming answers, and visible sources.