Vikram Setty

I'm

About Me

Vikram Setty

I'm Vikram, an ADAS Sensor Data Engineer at Lucid, working on enhancing the customer driving experience by building and deploying robust multi-modal machine learning solutions for ADAS/AD data flywheels.

Before my time at Lucid, I was a robotics graduate student at the University of Maryland (UMD), where I gained technical expertise and expereince in computer vision, robotics, reinforcement learning, multimodal and generative AI, and AR/VR technologies. During my time at UMD I was involved in a lot of research opportunities, and worked as a research assistant at the Perception and Robotics Group (PRG) in UMD headed by Dr. Yiannis Aloimonos where I deveopled autoencoder-based point cloud pretraining techniques. I also researched and implemented path planning algorithms for autonomous systems as a part of the Gamma Lab in UMD as well. During the summer of 2024, I also got the chance to intern at OmniVision, where I worked with event camera-assisted 3D reconstruction and spatial video generation. Apart from these experiences, I've also worked at the MarMot Lab in the National University of Singapore, MOIR Lab in IIT Ropar, and Aspiro, Bangalore where I was involved in similar research and projects.

I also work on various personal and academic projects in these areas and open-source my work on GitHub. I enjoy writing and frequently publish technical blogs on Medium as well.

In my free time, I love playing sports, and also love traveling, exploring new places, and meeting new people.

Skills

  • C & C++
  • Python
  • MATLAB
  • CMake
  • PyTorch
  • TensorFlow
  • CUDA
  • TensorRT & ONNX
  • ROS 1 & ROS 2
  • Gazebo
  • Isaac Sim
  • AWS
  • Azure
  • MySQL & PostgreSQL
  • MongoDB
  • RISCV-asm
  • Git
  • Linux
  • Docker
  • Kubernetes
  • Airflow
  • OpenCV
  • Open3D
  • Solidworks

Professional Background

Education

University of Maryland, College Park

Aug 2023 - May 2025 | College Park, MD

M.Eng. in Robotics

  • GPA: 4.0/4.0
  • Coursework:
    • Advanced Computer Vision
    • Multimodal Foundation Models
    • Robot Learning and Decision Making
    • Robot Perception and Planning
    • Robot Modeling and Control
  • Research: Graduate Research Assistant at the Perception and Robotics Group (PRG) and Gamma Lab

Indian Institute of Technology (IIT) Ropar

July 2018 - May 2023 | Ropar, India

B.Tech - M.Tech (Dual Degree) | Major: Mechanical Engineering | Minor: Computer Science

  • GPA: 8.97/10
  • Coursework:
    • Data Structures and Algorithms
    • Computer Architecture
    • Databases
    • Operating Systems
    • Applied Machine Learning
    • Computer Vision
    • Internet of Things (IoT)
    • Probability Theory
    • Differential Equations
  • Thesis: Mapping, Localization and Navigation of an Autonomous Robot Pursuing Dynamic Targets in Warehouse Environments
  • Awards: Institute Silver Medal for the highest GPA in the Mechanical Engineering (Dual Degree) cohort
  • Positions of Responsibility: Head Department Coordinator for the Career Development and Placement Cell (CDPC) from May 2021 - May 20-22. Led a team of about 20 people and connected with industry leaders, organized recruitment drives, and conducted professional development activities for students
  • Extracurricular Activities: Core Member of the Debating Society. Represented IIT Ropar as a debater and adjudicator in Inter IIT '18, '19, '23, and NSUT Colloquium '19.

Professional Experience

Lucid Motors

Jun 2025 - Present | Newark, CA

ADAS Sensor Data Engineer | ADAS GeoAI Team

  • Developing robust multi-modal machine learning and computer vision solutions for Lucid’s ADAS/AD data flywheels.
  • Deploying machine learning solutions for data auto-annotation for ADAS/AD BEV-Perception - project owner for VLM-based object attribute auto-labeling leading to a 95%+ speedup in annotation time.
  • Leading the data processing pipeline (from data collection to annotation), used for ML model development in Lucid’s auto-parking feature - have delivered over 150k frames of data for Lucid Gravity.
  • Enhancing sensor fusion, data synchronization, calibration, and logging capabilities on the test-fleet cars for in-house development and partner collaborations - scaled enhanced logging capabilities across 10+ data collection cars.

Gamma and PRG Labs, University of Maryland

Jan 2024 - May 2025 | College Park, MD

Graduate Research Assistant | Guides: Dr. Dinesh Manocha, Dr. Yiannis Aloimonos

  • Formulated and improved point cloud pretraining algorithms using VecKM-based autoencoder models.
  • Developed an autonomous navigation stack (with a custom local and global planner) for the Husky Robot to autonomously navigate campus environments through high-traffic conditions using camera, GPS, lidar, and IMU data.

OmniVision

May - Aug 2024 | Santa Clara, CA

Software Algorithm Intern | Camera Systems Group

  • Devised and developed novel Neural Radiance Fields (NeRF), Gaussian Splatting (GS), and Spatial Video pipelines for optimized rendering, view synthesis, and immersion from OVT image (CIS) and event camera (EVS) sensor data.
  • Researched, implemented, and benchmarked EVS 3D reconstruction algorithms including sparse multi-view stereo mapping, dense monocular depth estimation using recurrent CNNs, and dense ToF sensor-based depth interpolation.

MarMot Lab, National University of Singapore

May - Aug 2020 & 2021 | Singapore

Research Intern | Guide: Dr. Guillaume Sartoretti

  • Developed improved multi-agent pathfinding (PRIMAL) and search algorithms using visual attention and deep reinforcement learning.
  • Devised a smart traffic light system using transformers and an expert Conflict-Based Search (CBS) planner to improve NUS’ submission to AI Crowd’s Flatland Challenge.

Aspiro

May - July 2021 | Bangalore, India

Artificial Intelligence Intern | Software Development Group

  • Formulated a RASA Chatbot to handle chitchat and fallback cases using DietClassifier and TEDPolicy models.
  • Generated functionality for the chatbot to respond to closer and nudger statements using multi-intent classification.

Project Portfolio

PepeVLMify
PepeVLMify

Vision-language model (VLM) fine-tuning on custom annotated datasets using low-rank methods (LoRA, LoHa, LoKr)

CamDroneLoc
CamDroneLoc

Visual odometry for drones using a particle filter with image vector encodings as measurements

Structure from Motion (SFM)
Structure from Motion (SFM)

SFM 3D reconstruction using linear and non-linear triangulation, linear and non-linear PnP, and bundle adjustment

Accelerated RRT* - LDV
Accelerated RRT* - LDV

Optimized RRT* (in speed and performance) using low directional visibility

Stereo Vision
Stereo Vision

Depth map generation from an image pair using stereo rectification and disparity computation

AutoCalib
AutoCalib

Camera calibration using Zhang's Camera Calibration Algorithm pipeline

AutoPano
AutoPano

Automatic panorama stitching using traditional and deep learning-based homography estimation

Pb-lite Edge Detection
Pb-lite Edge Detection

Probability-based edge detection in images using brightness, color, and texture gradients from custom filter banks

Car Charging Robot
Autonomous EV Charging Robot

EV charging robot based on the FANUC CRX-10iA/L cobot with a custom perception module and inverse kinematics solver

Dynamic Fleet Management
Dynamic Fleet Management

Swarm robotics platform that incorporates multi-agent collision avoidance using a RVO-2 planner based on the ORCA algorithm

LQR & LQG Controller System
LQR and LQG Controller System

Custom LQR and LQG Controller to stabilize a two-pendulum cart system

Human Detection and Localization (HDAL)
Human Detection and Localization (HDAL)

Localizing the 3D coordinates of humans from a mobile robot at a fixed height using a YOLOv5 model

Minimum-Cost Arborescence
Minimum-Cost Arborescence

Implementation of Edmond's Branching Algorithm to calculate the minimum-cost arborescence of a directed and weighted graph

Heap Analysis - Johnson's Algorithm
Heap Analysis - Johnson's Algorithm

Time complexity analysis of array, binary, binomial, and Fibonacci heap implementations using Johnson's all-pair shortest path algorithm

RISC-V ISA Simulator
RISC-V ISA Simulator

Simulator (with GUI) that runs machine code for a 32-bit version of RISC-V Instruction Set Architecture

OS Algorithm Simulation
OS Algorithm Analysis

Set of C programs that simulate various algorithms and techniques used in general operating system design

Faculty Leave Portal
Faculty Leave Portal

University faculty leave portal with a MySQL and MongoDB backend and PHP frontend hosted on a local XAMPP server

Blogs

blog9
Generative AI in Reinforcement Learning

blog8
Event Cameras - Everything you need to know

blog7
Time Complexity Analysis of Dijkstra's Algorithm

blog6
Scene Representations in 3D Computer Vision

blog5
Using ROS in a Docker Container

blog4
Edge Types in Graph Traversal

blog3
Compiler Aware Programming Optimizations

blog2
The Bayes Filter

blog1
Solving Second Order ODEs Using Euler's Method & Python

Awards

Vikram Setty

IIT Ropar - Institute Silver Medal

April 2024

Awarded IIT Ropar's Institute Silver Medal in recognition of the best academic performance in the mechanical engineering dual-degree cohort.


Vikram Setty

KVPY Fellowship

April 2018

Awarded the prestigious KVPY SX fellowship by IISC Bangalore in recognition of extraordinary academic capabilities and research potential. Achieved an All India Rank (AIR) of 528 in the selection process for the same.

Hobbies

Vikram Setty

Outside of work, I'm a travel-hungry mid-twenties guy who wants to explore new places, try new food, go to concerts, play sports, and have a great time with my friends and family. I value experiences and want to do things, both professionally and non-professionally, that I enjoy, have fun doing, and grow from.

When I was younger, my family moved a lot, and I ended up growing up in a lot of different places, from India, to the UK, and the US. It's fair to say the traveller inside me didn't just awaken all of a sudden! Exploring the world, whether that be hiking in the national parks of the US, cafe-hopping in Japan, jet-skiing in Thailand, or looking down from the top of the London Eye, has broadened my view of the world, and brought a yearning to travel even more. Some of the things I want to do next on my bucket list are scuba-diving in the Great Barrier Reef, riding a hot air baloon in Cappadocia, and doing glacier treks in Iceland!

In general, I'm also an active person, and like playing sports and working out. I regularly play racket sports like tennis, badminton, pickleball, and ping-pong, and I love swimming too. I really enjoy and can also do elementary surfing and skiing, and am looking at doing a lot more of and getting better at these activities this year!

Like anyone else, I love listening to music. I follow a lot of genres, though my favorites are rock, pop, and indie. The past year has been extremely rewarding as I got to see some of my favorite artists live. Out of all of them, listening to Tame Impala live was probably my favorite! I also used to play the piano in middle school, but have been out of touch ever since then, although I did recently start playing the ukelele, and am learning how to play my favorite songs on it!

Another cool fact - I'm fond of and good at public speaking! I was a part of IIT Ropar's core debating team and have represented my university in numerous competitions (Inter IIT Cultural Meet '18, '19, '23, and NSUT Colloquium '19) as both a debater and adjudicator. I also love trivia, and was a part of IIT Ropar's Quiz Club during my undergraduate days.

I am a very social person and love meeting new people. Whether it's catching up with old friends for a game of Topgolf or solo traveling and meeting new people in a different country, I am always ready and embrace new experiences as they come! If you ever see me in person, please say hi! I'd love to chat!


Contact Information

Location:

San Francisco Bay Area, CA