About Me
A Bachelor of Engineering in Computer Science, I am a Data Engineer and Full-Stack Developer passionate about the lifecycle of data. From architecting scalable pipelines to crafting high-performance web applications, I focus on precision, performance, and clean code. I thrive at the intersection of data processing and modern web technologies, delivering end-to-end solutions that empower data-driven decision-making.
Architecting Data Foundations
I am a data engineer specializing in building exceptional digital experiences. Currently, I'm focused on building highly responsive, human-centered products.
My Works
Pulse Stream Analytics
An e-commerce observability project: a Python script simulates high-volume order/traffic data, DuckDB handles the analytical queries (fast, in-process, no separate DB server needed), and a Next.js dashboard shows live aggregations updating in near real time. Built to explore how far you can get with DuckDB before you actually need a "real" streaming stack.
Aura Lakehouse
A data lakehouse built on Apache Iceberg, following the medallion architecture (bronze → silver → gold). It supports time travel (querying data as it looked at a past point), schema evolution without breaking downstream consumers, and automatic PII masking. Paired with a Next.js dashboard so you can actually see the pipeline state, not just query it from a notebook.
Ledger Logic
Takes messy, ERP-style transaction logs from a simulated BNPL (buy-now-pay-later) business and turns them into a clean SQL star schema. Includes a metrics layer so KPIs are defined once and reused everywhere, automated checks to catch financial inconsistencies, and documentation written specifically so an LLM can reliably query the data without hallucinating definitions.
Sentinel Flow
A real-time pipeline (Kafka + Flink) that watches streaming data for schema drift and failures. I built an AI agent on top using LangGraph that reads the logs, figures out what broke, and suggests a SQL fix — with a human approving before anything runs. The goal was to see how far you can push "AI ops" without giving up control over production data.
Cloud Stream Analytics
A local sandbox that mimics AWS event-driven architecture without any cloud cost. LocalStack stands in for Kinesis, a FastAPI service consumes the stream asynchronously, and a Next.js dashboard shows live metrics over WebSockets. Built to practice event-driven design patterns before deploying anything to real AWS infra.
Let's build your next
masterpiece.
I'm currently available for freelance work and full-time opportunities. Have a project in mind? Let's talk!
Email Address
baazithh.p@gmail.com