Hello, I'm
I build backend systems and data pipelines — and on the side, I'm into ML and reinforcement learning, exploring more. I break things to understand them, debug what others skip over, and optimize until it's right. 8+ years shipping code across four industries.
the quick version
what I've worked on and what I took from it
Anuvu · Chicago, Illinois
Working on high-throughput data pipelines and real-time telemetry systems for in-flight connectivity. What I enjoy most here is the scale — 1TB of daily logs, 2,000+ flights — and the satisfaction of re-architecting the log-download pipeline so we don't quietly lose flight data anymore. Missing-log rate went from 15–20% to under 1%, and that one fix is what every other system upstream depends on.
Anuvu · Lombard, Illinois
Built monitoring, diagnostics, and data collection tools across avionics systems — wifi quality scoring, weather outage detection, modem failure diagnosis. I liked that the problems were novel; nobody handed me a playbook. I had to figure out why things were failing and build the tools to fix and prevent it.
TEOCO Corp · Overland Park, Kansas
Audited telecom billing circuits and built ETL pipelines to surface revenue leakage. The detective work is what I enjoyed — digging through contracts and data until the discrepancy made sense, which led to identifying over $1M in billing errors.
Foxbat Research · Dallas, Texas
Redesigned database architecture and built ETL pipelines for a finance application. A short engagement, but I came in, found what needed improving, and shipped it.
Foxbat Research · Dallas, Texas
Benchmarked transfer learning and sentiment analysis models across vision and NLP tasks. This is where my interest in ML properly started — hands-on with real models, real data, real tradeoffs.
Infosys BPM Ltd · Pune, India
Built full-stack web applications and end-to-end RPA automation for finance processes. What stuck with me was seeing how much manual work could be eliminated — 11 workflows automated, 75% reduction in handling time. That experience shaped how I think about engineering: if something is repetitive and painful, it's a problem worth solving.
side projects and things I got curious about
Multi-agent literary debate engine. Upload a book, ask a "what if" question, and AI versions of the characters argue out the alternate ending — moderated by an elephant, with per-character voices, an argument ledger, and a live force-directed interaction graph. Streams over SSE; runs on any single API key.
Travel companion app that connects travelers with helpful people on the same flight. Find a friend in the sky — get assistance, share rides, or just have company on your journey.
Interactive educational platform for learning data structures and algorithms through step-by-step canvas visualizations, multi-language code examples, and quizzes. 19 topics covering sorting, searching, trees, graphs, and algorithm techniques.
Chrome extension that analyzes job postings against your resume using AI — match scores, skill gap analysis, salary & location extraction, auto-fill applications, and applied jobs tracking.
Chrome extension that summarizes any webpage using your choice of AI provider — 13 providers including free local Ollama. Translates summaries into 84 languages, reads aloud with TTS, exports to PDF/Markdown, and saves to searchable history.
Agent that learns to navigate a grid with zero location info. Uses Q-Learning with Bayesian belief states to track probability of being in each cell under partial observability.
Computer vision model that detects chess pieces and locates them on the board. Combines image analysis (Hough transform, corner detection) with a deep neural network for piece classification and move tracking.
Statistical analysis of historical stock data with pattern visualization. Trained RNN and LSTM models to predict next-day stock prices from time series data.
tools I've actually used, not just heard of
where it started
The University of Texas at Arlington
2018 — 2020
University of Pune
2011 — 2015
always up for a good conversation
If you're working on something cool, I'd genuinely like to hear about it. New project, interesting problem, something you're building — reach out. Always up for a good conversation.