Param Desai
i do what i gotta do • B.Tech CSE, Nirma University • IIT Madras
Education
Nirma University, Institute of Technology
B.Tech in Computer Science and Engineering
Courses: OS, DSA, OOP, DBMS, Networks, DAA, SE, Distributed Computing, Cloud Computing, ML, Deep Learning, Full Stack, Compiler Design
Indian Institute of Technology Madras
Diploma in Data Science (In Progress) | Foundation & Programming Completed
Courses: ML Foundations, ML Techniques, ML Practice, Business Analytics, Data Science Tools, MAD I & II, DBMS, DSA
Research Experience
Research Assistant
Sudeep Tanwar's Research Group
- Undergraduate research focusing on Explainable AI (XAI), Adversarial Machine Learning, Cybersecurity, and IoT Systems under the guidance of Prof. Sudeep Tanwar and Prof. Rajesh Gupta.
- Designed and evaluated ML/DL-based intrusion and false data injection attack detection frameworks for connected vehicle environments.
- Applied model interpretability methods (SHAP, LIME) to analyze feature importances and audit decision metrics.
- Contributed to modeling, experiment validation, and publishing of 5 peer-reviewed papers at IEEE and Springer venues.
Industry Experience
AI/ML Intern
DYU Labs
- Technical lead for a summer intern team building production solar forecasting pipelines and a conversational analytics assistant.
- Developed end-to-end day-ahead time-series forecasting models using LightGBM for 75 production inverter devices across 15 solar plants.
- Designed conversational AI workflows leveraging multi-agent LangGraph architectures, RAG (pgvector), and natural language databases.
- Deployed automated ingestion and ML pipeline components on AWS (Lambda, Fargate, DynamoDB) for clients including ISRO, NTPC, and Aditya Birla Group.
Publications
- "FIDES: Federated Intelligence and Detection with Quantum Security for Financial Institutions." In Proceedings of the International Conference on Computer, Information and Telecommunication Systems (CITS 2026), 2026. (Accepted)
- "Machine Unlearning-based Privacy-First Medical Imaging Framework for TB Detection." In Proceedings of the Scientific Workshop on AI for Medicine (AI2M4RI 2026), 2026. (Accepted)
- "Q-PhishNet: Quantum-Secured Explainable Machine Unlearning for Phishing Detection in IoT Networks." In Proceedings of IEEE Globecom 2025, Taipei, Taiwan, 2025. DOI: 10.1109/GCWkshps68340.2025.11590920.
- "Q-ShielD: Quantum-Enhanced Secure Framework for Autonomous Vehicles Communication." In Proceedings of the IEEE 101st Vehicular Technology Conference (VTC Spring 2025), Oslo, Norway, 2025. DOI: 10.1109/VTC2025-Spring65109.2025.11174381.
- "Quantum-based Edge Intelligence Framework for Wearable Health IoT Device Networks." In Proceedings of the IEEE International Conference on E-health Networking, Application & Services (HealthCom 2025), Abu Dhabi, United Arab Emirates, 2025. DOI: 10.1109/HealthCom60686.2025.11342726.
- "Quantum-Assisted XAI-Driven DL Framework for FDI Detection in Connected Autonomous Electric Vehicles Underlying 6G." In Proceedings of the IEEE 5th Intelligent Cybersecurity Conference (ICSC 2025), Tampa, FL, USA, 2025. DOI: 10.1109/ICSC65596.2025.11139950.
- "Explainable AI and Quantum Security for Smart Homes Network Attack Classification." Springer Journal of Computational Machine Learning (CML 2025), 2025.
Projects
- BHOOMI — An environmental intelligence platform integrating satellite datasets with multi-horizon forecasting models for climate monitoring and AI-assisted municipal reporting.
- FC Analytics — A full-stack sports analytics platform featuring real-time match forecasting, Monte Carlo tournament simulations, and live multi-user auction systems.
- PetBot — An LLM-powered RAG system with natural language to SQL capabilities, revenue analytics pipelines, and voice-assisted conversational workflows.
- Research Assistant — An offline, local-first paper assistant that automatically extracts and processes PDFs to generate section-aware summaries, simplified drafts, PPT decks, and podcast audio.
- Developer Portfolio — An interactive, high-performance academic and engineering portfolio website featuring smooth dark mode transitions, a custom Markdown blog, a command palette, Supabase realtime guestbook components, and reactive theme-based diagram rendering.
Achievements
- Runner-Up, HACKaMINeD 2026 (PetBot)
- Runner-Up, AETRIX 2026 (BHOOMI)
- Conference Presenter, IEEE ICSC 2025 — Tampa, Florida, USA
- Gold Medalist, SOF International Mathematics Olympiad (IMO)
- Silver Medalist, SOF National Science Olympiad (NSO)
- National-Level Representative, Spell Bee Examination
- Distinction in Communication Skills (Grades 1–3), Trinity College London
- Represented Ahmedabad International School at MatheGon, IIT Gandhinagar
- Representative at the PYP Exhibition
- Cleared A1 and A2 levels of French Language Examination
Leadership
Treasurer, CSI Nirma University
- Managed financial strategies and execution for HACKaMINeD 2026 (2500+ registrations, INR 4.5L prize pool, held in collaboration with SUNY Binghamton).
Core Committee Member, CSI
- Organized national-level hackathons and tech symposiums including HackNUThon 6.0 and CUBIX 2025.
Skills
Frameworks
FastAPI FastAPI React React LangChain LangChain LangGraph LangGraph Flask Flask Express.js Express.js Vite Vite
ML/AI
Monte Carlo (MC)RAGNLPProphetSARIMASQLAlchemy LightGBM LightGBM
Tools
PostgreSQL PostgreSQL DynamoDB Docker Docker Git Git AWS ECR & LambdaWebSockets Supabase Supabase
Research Areas
XAIAdversarial MLMachine UnlearningQuantum Security
Community Service
- Volunteered in teaching and mentoring underprivileged children.
- Participated in educational outreach and community engagement activities.