I'm Aditya Sidham. Computer vision and applied AI are what I love working on, because the interesting part starts when a system stops recording the world and starts understanding it. I've turned that into a patented form‑correction coach, a four‑agent nutrition app, and a tool that lets anyone query a database in plain English. My ambition is simple: ship AI products people actually rely on, and keep getting better at the craft.
Selected work
My projects start with a practical question. Can a camera coach exercise form? Can a database understand someone who doesn't know SQL? Can nutrition tracking respond to real behaviour? I answer by building the full loop, then fixing the parts that break in real use.
Flagship AI + native iOS · 2026
A nutrition app that learns from what you actually do. Solo‑built: the SwiftUI app, the FastAPI backend, the OCR pipeline and a four‑agent Claude layer that turns a label photo into a plan you can argue with.
Real‑time exercise feedback from an ordinary camera. Tracks 33 body landmarks, counts reps only when the form is right, and led to a published patent.
Ask a database a question in plain English. Schema‑aware prompting keeps the query grounded, a review step keeps the human in control. Roughly 60% faster query construction.
A fire alarm that can reach you when you're not there. ESP8266, flame and smoke sensors, alerts over GSM and Wi‑Fi.
Computer vision that produces a physical response: count fingers on camera, light the matching LEDs through an Arduino.
The project that made engineering feel real: a multi‑axis servo arm, later driven by gyroscope hand gestures.
About
I've never been satisfied knowing only the part of a system assigned to me. If I'm using a tool, a sensor or a model, I want to know what it's doing, where it fails and how it fits into everything around it. That instinct has taken me across a wide range: Arduino circuits, computer vision, enterprise backends, native iOS and multi‑agent AI systems.
My path started with physical computing, debugging with a multimeter as often as a terminal. Computer engineering gave me the software foundations. At TCS and LTM the code sat inside larger operational workflows, so correctness and reliability mattered. At USC I've gone deeper into machine learning, data mining and language systems.
FitandFine brings those threads together: a product shaped by something I care about personally, where I own the interface, the backend, the data model, the OCR pipeline and the AI layer at once. That kind of ownership forces me to think about the user and the architecture at the same time, and it's the work I want to keep doing.
Documentation is a starting point. A working prototype tells me what the real problem is. I get something running early, inspect each layer and improve it with evidence.
A model is only useful if the data reaches it reliably and the result makes sense to a person. I care about the whole path from input to interface.
Caching, validation, error states, background jobs and offline behaviour decide whether a demo becomes a dependable product. I enjoy that work.
I'm comfortable owning a problem, but independence isn't silence. I clarify assumptions, share progress and surface risks before they get expensive.
Off the clock
Training made me patient with slow progress and honest about feedback. You can't talk your way around a missed workout or a weak system. You put in the work, measure what happened, and adjust. It's the same loop I run on code.
Follow the training log
Experience & education
Large delivery teams and small hands‑on projects. Different environments, same expectation: understand the system, make the change carefully, leave it working better than before.
University of Southern California Los Angeles, CA GPA 3.52
LTM, formerly LTIMindtree Mumbai, India
Tata Consultancy Services Navi Mumbai, India
DY Patil Ramrao Adik Institute of Technology Nerul, Navi Mumbai GPA 3.68
Aces Invensys Technologies Navi Mumbai, India
Highlights
A published patent, a flagship product, a national finals, a maker award, and the certificates that taught me to connect technical decisions to business context.
Application 202421099595, "AI Powered Exercise Tracking and Form Correction System", published 17 January 2025. Real‑time pose landmarks that count reps and check form through a camera.
Label OCR to structured macros at over 85% field accuracy, weekly diet analysis, meal recommendation and a closed‑loop calorie target, all behind a native iOS app.
Led a four‑person team that built SafeSense, an ESP8266 smoke and fire detector with local alarms plus GSM and Wi‑Fi notifications. Selected as a Creative Ideas & Innovations in Action finalist.
Recognised for a servo‑driven robotic arm I designed and programmed, later extended with gyroscope‑based gesture control.
Data preparation, analysis, visualisation and communicating findings, completed on Coursera.
Business Analytics and Sustainable Business Strategy. The programs taught me to connect technical decisions to evidence, business context and long‑term trade‑offs.
Contact
I'm interested in applied AI, software engineering, data and product work where I can take real ownership and keep learning. If you're building something that has to work beyond the demo, tell me what you're trying to solve.
sidham@usc.edu
