Hey there! I'm a Computer Science graduate from VIT Chennai with a passion for distributed systems, federated learning, and high-performance computing. What started as academic curiosity has evolved into building production-scale systems that solve real-world problems.
I'm particularly drawn to challenges involving distributed machine learning, edge computing, and systems that need to scale efficiently. My recent work focuses on federated learning frameworks and containerized ML pipelines that can run across multiple devices and environments.
What I'm Currently Working On
Current focus areas:
- Leading the development of HabitatFL, a federated multimodal learning framework for wildlife conservation
- Building enterprise-grade banking systems with modern web technologies
- Expanding GreyLattice publication with deep-dive technical content on neurotechnology
- Exploring the intersection of edge computing and machine learning
Recent achievements:
- Architected federated learning systems achieving 87.65% accuracy on wildlife audio classification
- Published peer-reviewed research on cryptographic pipelines and adaptive cybersecurity
- Deployed containerized ML workflows on NVIDIA DGX infrastructure
- Founded and launched GreyLattice technical publication with 5+ longform articles
|
Leading HabitatFL project - a federated multimodal learning framework achieving 87.65% central accuracy on wildlife classification. Built lightweight edge-compatible models (<2MB) that outperform centralized training by 5%+. Tech Stack: Flower, Docker, NVIDIA DGX, MobileNetV3, SqueezeNet Impact: Enabling privacy-preserving ML for wildlife conservation across distributed edge devices. |
Co-authored research on JIGS cryptographic pipeline achieving 96.67% speedup through parallelization. Designed Siren adaptive cybersecurity framework with ML-driven threat engagement. Tech Stack: ECDH, AES-128, Python, TCP sockets, ProVerif Impact: Published peer-reviewed papers on modular encryption and deception-based security. |
|
Independent Project Lead (Jan 2025 - May 2025) Architected a federated multimodal learning framework for wildlife audio classification and visual recognition. Achieved 87.65% accuracy on 18.2K spectrograms (27 classes) and 69.92% on 50K images (82 classes) using lightweight edge models.
Key Achievement: Outperformed centralized training by 5%+ while maintaining edge compatibility Impact: Enabling privacy-preserving ML for distributed wildlife monitoring systems |
Full-Stack Development (May 2025 - Present) Building a production-ready banking system with FastAPI backend and React frontend. Features role-based access control, ACID-safe fund transfers, and comprehensive testing with 85+ backend coverage.
Key Features: JWT authentication, SQLAlchemy ORM, Docker containerization, CI/CD pipeline Focus: Learning enterprise-grade software architecture and modern web development practices |
|
High-Performance Computing Project (Jan 2024 - Apr 2024) Built a real-time financial sentiment analysis pipeline processing high-volume news streams. Reduced latency from 8 minutes to 10 seconds using RAPIDS cuDF and Apache Kafka on RTX 2060.
Performance: 48x speedup through GPU acceleration and multiprocessing optimization Learning: Real-time stream processing, GPU computing, and containerized deployment |
Research Project & Publication (2024) Designed and implemented a modular parallel encryption system achieving 96.67% speedup. Co-authored peer-reviewed paper published in IOP SciNotes with formal verification using ProVerif.
Research Impact: Published methodology for parallel cryptographic operations Technical Achievement: Formal security verification and performance optimization |
|
Time Series ML Project (Jan 2024 - Apr 2024) Built 3-layer LSTM model achieving 92% accuracy in predicting NAVs across 50+ mutual fund portfolios. Implemented walk-forward validation and backtesting framework for investment simulation.
Techniques: Sliding window preprocessing, Min-Max scaling, time-aware validation Application: Production-ready investment decision simulation framework |
Google Developers Externship (Aug 2023 - Nov 2023) Developed YOLO-based weapon detection system with Flask web interface. Optimized for real-time video stream processing with live alerting capabilities.
Team Collaboration: 3-member distributed team across design, development, and presentation Deployment: Live web application with real-time webcam stream processing |
- JIGS: A Modular Parallel Encryption and Authentication Pipeline for Secure Messaging (2024)
Published in IOP Engineering Research Express - Researched on parallelized cryptographic operations achieving 96.67% speedup
Link to paper - Siren: Adaptive Cybersecurity through Deception and ML-Driven Threat Engagement (2024)
Peer-reviewed preprint - Designed deception-based framework with ML classification and honeypot integration
Link to paper
Founded and launched GreyLattice - a technical publication focused on neurotechnology and computational neuroscience. Produced 5+ longform technical articles reaching technical audiences interested in brain-computer interfaces and neural engineering.
- GL00: GreyLattice — The Next Revolution in Human-Machine Interface
Vision piece outlining the platform’s mission and framing the neurotechnology landscape - GL01: The Foundations of Computational Neuroscience
A 25–30 min read exploring the evolution of computational neuroscience and key theoretical developments - GL02: How We Listen To The Brain — And Why It’s So Hard
Deep dive into invasive vs non-invasive brain recording methods, spatial-temporal tradeoffs, and signal degradation challenges - GLr01: Brain Decoding — Toward Real-Time Reconstruction of Visual Perception
50-minute technical breakdown of MEG-based decoding models, visual cortex reconstruction, and neural representation learning (ArXiv:2310.19812) - GLc01: Podcast Pilot Episode
Narrative-driven podcast episode introducing GreyLattice through story, philosophy, and deep tech insight
Independent Project Lead - HabitatFL (Jan 2025 – May 2025)
Remote Access via VIT NVIDIA DGX Infrastructure
- Led development of federated multimodal learning framework for wildlife conservation
- Achieved 87.65% accuracy on audio classification and 69.92% on visual recognition tasks
- Deployed containerized Flower-based pipeline orchestrating multi-GPU CUDA workflows
Research Assistant - Secure Systems & AI (Jan 2024 – May 2024)
Vellore Institute of Technology
- Co-authored 2 peer-reviewed papers on encryption pipelines and adaptive cybersecurity
- Built Python-based cryptographic systems with ECDH and AES-128 implementation
- Designed ML-driven honeypot systems for proactive threat engagement
Data Science & ML Extern (Aug 2023 – Nov 2023)
Google Developers x SmartBridge (Remote)
- Developed real-time weapon detection system using YOLO and Flask
- Optimized TensorFlow models for real-time webcam stream processing
- Collaborated with distributed team on end-to-end system design and deployment
Founder & Technical Writer - GreyLattice (Feb 2025 – Present)
Independent Technical Publication
- Founded neurotechnology publication with 5+ longform technical articles
- Produced content on neural recording, MEG-based models, and brain-computer interfaces
- Launched pilot podcast episode focusing on narrative-driven tech communication
I'm actively seeking opportunities where I can:
- Apply federated learning and distributed ML to real-world problems at scale
- Work with high-performance computing systems and learn from experienced platform engineers
- Contribute to edge AI and privacy-preserving ML systems in production environments
- Collaborate with cross-functional teams on challenging technical problems
Current interests:
- Federated learning infrastructure and optimization
- Edge computing and mobile ML deployment
- Distributed systems and containerized ML workflows
- Applied cryptography and secure multi-party computation
I'm passionate about discussing federated learning, distributed systems, and the intersection of privacy and machine learning. Always excited to learn from others and share experiences from my projects.
|
|
|
|
Based in Chennai, India (UTC +05:30) • Open to Remote & Hybrid Opportunities