I thrive at the intersection of hardware and software, with experience in embedded systems, robotics, controls, and computer vision. I'm actively seeking interesting, multifaceted roles in these areas.
⤢ Closer look
I'm an ECE Honors undergraduate at UT Austin who likes problems that don't fit neatly into "hardware" or "software." I've written OS-level drivers at Qualcomm, shipped flight-control, maintenance, and testing software at Bell Flight, built a LiDAR SLAM localization stack with the UT Dallas NOVA team, and architected prototyping, testing, and control solutions at Texas Guadaloop.
What drives me is seeing my work come to life in realistic scenarios. I've done this through building closed-loop control systems on real hardware, bringing up computer vision systems that function in practical spaces, and working on multiple sensor-fusion projects that solve real problems. I care about systems that are fast, robust, and honest about their limits.
Tap any skill to see the projects and courses where I put it to use.
A mix of research, internships, and hackathon wins. The Guadaloop rig and RescueVision both have live, scroll-driven 3D walkthroughs.
3D
A magnetic levitation test rig I lead development on. Built the control stack from hand-tuned PID up toward a FEA+PyBullet sim-based pipeline. Working to make levitation scalable.
Open the 3D walkthrough →
3DAn mmWave + UWB system that locates victims through walls and renders them in a first responder's own AR frame. Won Analog Devices' Sensor Fusion track at Purdue's StarkHacks.
Open the 3D walkthrough →
3DA PT app fusing CV-based pose estimation with an optional 2-IMU brace via an EKF for recovery tracking. Won the Healthcare track at CMU's Nexhacks.
Open the 3D walkthrough →Full-stack IoT smart blinds — XIAO ESP32-C6 using FreeRTOS, Express backend, and a Flutter app, with an end-to-end user experience.
Open the 3D walkthrough →
An autonomous racer on a custom RTOS — PD wall-follower with optional RL residual control + custom heuristic-based overtaking state machine. Won the ECE445M final race. (I developed the PD controller and classical logic, and we won without an RL model 👀)
Read case study →
A LiDAR SLAM + map-based localization pipeline for UT Dallas NOVA's autonomous vehicle, built on KISS-ICP / Open3D. Fused ICP with feature matching for sub-5 cm accuracy in CARLA.
Read case study →
A browser-based instrument simulator with hand tracking so you can see and hear yourself play (even learn!) without buying or making a trip to a store. "Most Risky Hack" at the TVG Vibeathon.
Read case study →Embedded software, robotics research, and controls — across industry, university labs, and student engineering teams.
Windows on Snapdragon. Extending device drivers to handle multiple clients and building a testing framework for the sensor suite. Driver/OS-level work.
Built ROS2 system using the Spinnaker SDK to pull frames over limited bandwidth from FLIR Blackfly S GigEVision cameras, correlate keypoints from multiple views, and output global pose estimates for people and obstacles in a space. Now working toward a markerless multi-camera calibration pipeline using humans as calibration targets.
Leading development of our control-algorithm and a repeatable formula for the HEMS levitation success.
Resolved 5+ software problem reports across FCS code and low-level tests; built a Python tool that migrated ~1000 high-level tests and comms databases between environments, saving an estimated 6–12 months of manual work. Resolved 13 key issues with .NET windows app, gaining comfort with .NET C# code using WPF.
Built the LiDAR SLAM + map-based localization stack (Open3D / KISS-ICP), reaching sub-5 cm localization accuracy via combined ICP and feature matching.
Monocular localization pipeline for the Sentry robot using OpenCV SolvePnP on standardized markers; tested odometry methods for tighter positioning including dead-wheel and encoder.
Product-dev intern on the KneeStim rehab device; circadian-rhythm biochemistry research; and co-captain of a state-championship-debut FTC robotics team.
"Aditya is a pleasure to work with and has excellent engineering skills… he developed full software-lifecycle solutions for our flight control computer software, exhibited agile thinking, and took initiative to suggest and implement additional improvements."
— Steve Feenstra, Flight Controls Software Lead / Staff Software Engineer, Bell Flight
ECE Honors coursework at UT Austin — embedded systems, computing, signals & circuits, and the underlying math.
Foundational calculus, engineering physics, and university core requirements were cleared through AP and transfer credit (26 hours).