Naghul Adhithya Venkateswaran
CS & Data Science @ UIUC
Focused on machine learning & high-performance systems across GPU-accelerated computing, low-latency execution, and distributed systems.
About
Building systems that are fast, efficient, and built to scale.
I build high performance systems with experience spanning machine learning and software engineering. My interests focus on GPU systems, low latency computing, and machine learning for scalable infrastructure.
I enjoy understanding how systems operate internally and building infrastructure that improves performance, efficiency, and scalability.
Outside of technology, I lead teams, stay driven through running and lifting, document the journey, and capture perspectives through photography.
Education
University of Illinois Urbana–Champaign
B.S. in Data Science & Information Sciences · Minor in Computer Science
Expected May 2029
Champaign, Illinois
Experience
Software Engineering Research Intern
The Ohio State University
May 2025 – Sept 2025
Columbus, OH
- Engineered benchmarking pipelines evaluating human vs LLM-generated object-oriented systems across programming workflows.
- Analyzed abstraction quality, modularity, and reasoning efficiency in collaborative human–AI software engineering paradigms.
- Developed reproducible evaluation frameworks for cognitive variation analysis in generative programming environments.
Machine Learning Systems Intern
Delhi Technological University
May 2025 – July 2025
Remote
- Designed multimodal biometric authentication pipelines integrating keystroke dynamics, achieving 99.47% classification accuracy.
- Optimized deep learning feature extraction using metaheuristic algorithms, improving cross-modal verification scalability.
- Evaluated distributed identity inference workflows under heterogeneous conditions for reliable real-time authentication.
Generative AI Engineering Intern
Indian Institute of Information Technology, Tiruchirappalli
May 2024 – July 2024
Remote
- Fine-tuned FLUX.1-dev diffusion architectures with LoRA adapters for controllable sketch-to-image generation rendering.
- Optimized latent diffusion and prompt-conditioning workflows, improving texture preservation and generative consistency by 20%.
- Engineered style-transfer pipelines enabling scalable prompt-guided colorization and artistic image synthesis.
ML & Data Analytics Research Intern
Incognito Blueprints
May 2023 – Aug 2023
San Francisco, CA
- Designed ML pipelines with ModelOps principles, boosting throughput 25% for 20+ SMB clients using MySQL.
- Optimized decision systems with real-time inference and performance analysis, improving accuracy by 20% and reducing lag by 30%.
- Published reproducible research enabling scalable experimentation and integrated NLP workflows into distributed ML systems.
Projects & Research
Springer · ICMEET
Real-Time Driver Drowsiness Detection
Deep learning-based driver monitoring using Eye Aspect Ratio, facial landmarks, and HSV-based sunglass detection for safety-critical transportation environments.
Springer · ICDSA
Real-Time Vehicle Detection & Counting
YOLOv8-based traffic analytics pipeline for real-time congestion monitoring with centroid-based multi-object tracking.
Team Lead
Ethical AI Vision System
Led a four-member global team developing AI-powered driver monitoring achieving 95% detection accuracy. Integrated OpenCV and TensorFlow pipelines for real-time eye tracking with fairness and bias mitigation review.
Skills
Languages
Frameworks
Infrastructure
Developer Tools
AI / ML
Systems
Contact
Let's build something.
Open for internships, research collaborations, and thoughtful conversations.