I design the data infrastructure that makes AI systems intelligent — RAG pipelines, vector search, and LLM integrations grounded in reliable, governed data. 5 years in finance-domain data engineering. Databricks Certified GenAI Engineer.
LLMs are only as good as the context they receive. I build the data layer that grounds AI agents in accurate, governed, domain-specific knowledge.
Retrieval-Augmented Generation architectures on Databricks — chunking, embedding, retrieval and re-ranking for finance-domain documents.
Databricks Vector Search & Unity Catalog for governed, low-latency semantic retrieval over large document corpora.
LLM chains and agent orchestration using MLflow, Model Serving, and structured tool-use patterns for data-grounded AI.
A focused selection of implemented systems. Each case highlights a concrete engineering problem, the design decisions behind it, and evidence from the working implementation.
Typed long-term memory for AI agents with temporal validity, contradiction and supersession handling, governed writes, memory packs, receipts, and JSONL traces. Implemented with DuckDB, provider adapters, a CLI, an authenticated HTTP API, Docker, and evaluations.
An auditable Raw → Silver → Gold pipeline for RSS, YouTube, and GitHub sources. Prefect coordinates deterministic ingestion and validation with concurrent agent analysis, source routing, synthesis, and delivery to a personal knowledge system.
An enterprise operations variant using Azure AI Foundry agents, Azure AI Search, evaluators, OpenTelemetry, Bicep infrastructure, and GitHub Actions quality gates. The implementation is being prepared as a clean standalone case.
Nine source adapters feed a shared event schema with source precedence, entity resolution, deduplication, geocoding, administrative corrections, and a public map application.
A typed production pipeline for scripts, audio, visuals, overlays, and final video assembly. Explicit draft, QA, approval, and final states keep human review inside the workflow.
A published hackathon implementation with Microsoft Agent Framework agents, typed Silver data, analyzer and synthesizer stages, Teams delivery, and three FastMCP services for blog, GitHub, and YouTube sources.
Formal certifications and ongoing hands-on achievements.
Design and implement an MLOps and GenAIOps infrastructure, manage machine learning model lifecycles, implement generative AI quality assurance and observability, and optimize generative AI systems and model performance.
Design and implement LLM-enabled solutions, RAG applications, and LLM chains using Databricks Vector Search, Model Serving, MLflow, and Unity Catalog.