Multi-Agent Architecture
Supervisor, hierarchical, and parallel fan-out patterns in LangGraph, with structured output you can actually trust.
AI Engineer
Systems, whisked to production grade.
I build multi-agent AI systems, LangGraph orchestration, RAG pipelines, and QC automation that hold up outside a notebook. Slow iteration, careful evaluation loops, steady results.
2+Years in AI systems
AWSAI Cloud Engineering
EN / VIBilingual delivery
About
Careful with the details that matter.
I spend most of my time in the space between a working prototype and a system a team can actually rely on: deployment, evaluation loops, retry logic, schema validation, and the plumbing that keeps agents honest.
RAG Pipelines
Retrieval systems with LangChain, Pinecone, Chroma, MMR, and hybrid search tuned for real documents, not benchmark demos.
Cloud & Ops
FastAPI, Docker, GitHub Actions, AWS learning, and secure deployment habits for AI systems that need to ship.
Selected Work
A few systems I have shipped.
Matcha
A local-first AI CLI tool with a LangGraph agent runtime, swappable storage backends, and a persistent memory, skills, and tools directory structure.
LangGraphCLILocal-first
Go to repositoryVi Recruiter
A Flask-based talent acquisition assistant with parallel CV processing, multi-platform candidate search, and PostgreSQL-backed memory.
FlaskPostgreSQLDocker
Go to repositoryAWSLangGraphLangChainPineconeChromaFastAPIFlaskDockerGitHub ActionsPyTorchPostgreSQLStreamlit
Contact
Have a system that needs building?
Open to freelance and collaboration on AI product work.
Email: nicole.ai@nicolethedev.top