// Hello, World
I'm Akshat Kumar Jha
Data Science & AI Engineer with production experience in LLM fine-tuning, multi-agent system design, and full-stack development. Building scalable ML pipelines on GPU infrastructure from prototype to production.
// Hello, World
Data Science & AI Engineer with production experience in LLM fine-tuning, multi-agent system design, and full-stack development. Building scalable ML pipelines on GPU infrastructure from prototype to production.
Final-year B.Tech student at VIT Vellore with a CGPA of 8.81, specializing in Electronics & Communications Engineering. I bridge the gap between hardware and intelligent software, from writing embedded firmware for IoT products to fine-tuning large language models on NVIDIA DGX infrastructure. Experienced in building end-to-end ML pipelines processing millions of records using PyTorch, RAG, and QLoRA, and delivering production REST APIs and real-time dashboards that serve thousands of claims.
Applied 4 ML models (DQN, Double DQN, Fuzzy Logic, Reinforcement Learning) to optimize spectrum allocation in Cognitive Radio Networks under Prof. Dr. Shanidul Hoque. Reduced handoff latency by 20% across 1,000+ simulated events. Paper submitted to Scientific Reports (Nature).
Designed a motion-activated wheelchair prototype using the MPU 6050 sensor and Embedded C control algorithms, achieving sub-100ms response time for real-time adaptive mobility.
Trained a Convolutional Neural Network from scratch using TensorFlow and Keras, achieving 82% accuracy in multi-category image classification with a full data pipeline.
Investigated advanced TFET design concepts under Dr. Shanidul Hoque, proposing 2 novel approaches to improve device efficiency by 12%.
Kaggle × Google · 2023
Cognizant · 2023
Maven Silicon · 2024
Boston Consulting Group (BCG) · 2023
I'm always open to discussing new opportunities, interesting projects, or collaborations in AI/ML, embedded systems, or full-stack development.