Apr 2025 – Jan 2026 · 9 mos · Hybrid

• Built an agentic CVE analysis platform that scans 1000+ ASTRA reports, identifies high-risk vulnerabilities, and auto-generates accurate remediation plans, reducing manual triage effort by 70%
• Developed LLM-powered code quality pipelines using prompt engineering, LoRA/QLoRA fine-tuning, and MISRA compliance checks, improving code maintainability and readability by ~30%
• Enhanced AoA detection in ISAC systems using OMP and VAE-based ML models, achieving 100% OMP convergence

Nov 2023 – Mar 2025 · 1 yr 4 mos · Hybrid

• Spearheaded development of enterprise-scale RAG system for 3GPP specifications using LangChain, and Mistral, achieving 92% accuracy in technical query responses surpassing ChatGPT 3.5's performance
• Deep Unfolded ISTA algorithm using Python, for the task of signal recovery from it’s compressed form. Achieving 10x faster convergence and 85% reduced computational complexity compared to traditional methods
• Refined a Vision Transformer model in PyTorch for semantic communication, optimizing
• bandwidth utilization by 45% and improving encoding efficiency by 30%. Leveraged attention mechanisms to emphasize critical image regions, ensuring high-fidelity reconstruction in low bandwidth environments.

Jan 2023 – Oct 2023 · 9 mos · Hybrid

• Implemented Convolutional Neural Networks for the task of Image Classification
• Implemented Physical Uplink Shared Channel till modulation for a client by writing CUDA kernels leading to a 27x faster performance gain via GPU parallelisation
• Implemented CUDA libraries like cuDNN for optimized Machine Learning models, cuRAND for generating random inputs, and Thrust for enabling performance portability between GPUs and CPUs

May 2021 – Jul 2021 · 2 mos · India

• Performed Exploratory data analysis on a dataset of over 50,000+ customers and used visualization tools to create actionable insights
• Trained an XGBoost model to identify at risk-customers and churn rates, selected optimal hyperparameters using GridSearchCV
• Used statistical analysis and A/B testing to determine the popular courses and optimal pricing strategies to maximize revenue

May 2020 – May 2021 · 1 yr · IIT Kharagpur

• Motivated and built a strong team of 19 Members to cover operations in Gwalior and its neighboring villages and cities
• Collaborated with 5+ schools of Gwalior and successfully organised a National Webinar under the initiative - RISE (Reform and Innovate School Education) which aims at promoting student-centric teaching methods
• Individually pitched maximum individuals among all LiGHT Centre for participating as mentors in Nation - Wide Career Counselling program which aims to provide indirective counseling to students of Secondary and Senior Secondary Classes