Aakash Madabhushi
I build the pipelines that feed dashboards and models, and the checks that keep bad data from reaching either.
A year and a half at Sapaad building production pipelines on Databricks and PySpark: more than five million rows a day across a multi-tenant SaaS platform, feeding recommendation features and over 75 tenant dashboards. That included tuning ten or more Spark jobs for a 20% cut in dashboard query latency, and a Delta Lake data quality framework with schema validation and anomaly detection, because those pipelines fed production ML.
The same discipline runs through my own work: an Airflow, dbt, and Snowflake medallion pipeline that reduced 27.1 million Zeek network flows into a graph a model could train on; a taxi platform built twice, once as a batch warehouse and once as a streaming pipeline with a dead-letter topic; and pre-training quality gates in my research that refuse to hand a corrupted tensor to a training run.
Every link goes to code or data you can inspect yourself.
The same taxi data handled two ways: a scheduled pipeline that builds dashboard-ready tables joined with weather, and a streaming one that validates trips as they arrive and publishes rolling five-minute metrics.
Personal project · Nov to Dec 2025
Trained a delay predictor on 9.5 million U.S. flights, then ran it inside a live stream so each incoming flight gets a delay probability within seconds. Where a flight lands and when it leaves drove most of the signal.
SJSU big data course, team project · Dec 2025
Streams every English Wikipedia edit as it happens and scores the tone of each edit summary, so you can spot pages turning contentious in real time.
Pulls daily stock prices into a warehouse, computes the indicators traders actually use, and keeps a full history of every revision to the data.
A scheduled pipeline that converts articles into vectors and makes them searchable by meaning rather than exact keywords.
A Yelp-style application rebuilt so services communicate through events instead of calling each other directly, letting each part fail and recover independently.
A LinkedIn-style hiring platform running as 17 coordinated services, plus an AI assistant that takes a recruiter from résumé to outreach and pauses for human approval before it sends anything.
SJSU DATA 236, Distributed Systems · Spring 2026
Condensed 27 million raw network connection logs into a map of which machines talked to each other, flagged the suspicious pairs with a graph neural network, and had a language model explain every alert in plain English.
SJSU capstone, team project · Feb to May 2026
The parts of my roles that bear on this work. The résumé has all of it.
Mar 2026 to present
Jun 2023 to Oct 2024
Sep 2019 to Mar 2020
I'm available from December 2026 and open to relocating. Email me and I'll reply within a day.
Hiring for something else?