Lead AI / Data Engineer · Building multi-tenant data platforms and production LLM systems
I'm a data and AI engineer with 6+ years of experience taking systems from whiteboard to production. I currently lead AI, backend, and data engineering at Descasio, where I'm building a multi-tenant Data Intelligence Platform on AWS (EventBridge, Glue, Athena, Lambda, ECS Fargate, Bedrock) that ingests, transforms, and serves analytics and LLM-powered insights across tenants.
- 🚀 Founder of MarketCore, an AI-native, multi-tenant marketing automation engine
- 📦 Previously led data engineering at Circo, designing event-driven engagement and ledger systems on Kafka, FastAPI, and dbt
- ✍️ I write about data engineering and ML on Medium and Hashnode
- 🤝 Open to collaborating on data platforms, applied LLM systems, and open-source tooling
| Area | Tools |
|---|---|
| Cloud & Infra | AWS (Glue, Athena, Lambda, ECS Fargate, EventBridge, S3, DynamoDB, Bedrock), GCP (BigQuery), Terraform, Docker |
| Data Engineering | Airflow, dbt, Spark, Kafka, SQL, Pandas, Soda |
| AI / ML | LangGraph, LangChain, RAG & GraphRAG, scikit-learn, computer vision |
| Backend | Python, FastAPI, Django REST, PostgreSQL, MongoDB, microservices |
| Project | What it does |
|---|---|
| data-contract-validator | Open-source CI guard that validates dbt models against FastAPI/Pydantic schemas, catching breaking contract changes before they ship |
| Enhanced GraphRAG | Knowledge-graph generator extending the GraphRAG-SDK framework to handle PDFs via Unstructured-IO |
| Airflow + dbt + BigQuery Pipeline | End-to-end orchestrated data pipeline with Airflow, BigQuery, dbt, and Soda data-quality checks |
| Audio Transcription System | Production-style audio transcription tooling built for scale |
| Enhanced Spark Modelling | Scalable Apache Spark solution for processing and analyzing large-scale task log datasets |
⚡ Off the keyboard: drawing, volleyball, mentoring beginners into data careers, and making music.