Maulik Chhetri

1. Designed a novel agentic-QA framework for Simulink models using LlamaIndex and RAG, achieving 90% accuracy on domain-specific engineering queries. 2. Deployed the full system (FastAPI backend, React frontend, Docker) and ran beta testing with 10+ Infineon engineers, collecting and incorporating iterative feedback. 3. Conducted a systematic comparison of 6+ LLM-based QA architectures (naive RAG, agentic RAG, tool-calling agents etc), using evaluation frameworks like LLMEval and RAGAS. 4. Collabor

1. Tackled expensive and time-consuming recording with RADAR camera, by developing data augmentation pipeline for a CNN-based RADAR presence detection system. 2. Improved a 3D CNN for RADAR-based presence detection in collaboration with the external RADAR team and ML team, achieving detection accuracy above 90% across all tested scenarios. 3. In order to understand the complex inner-workings of the presence detection model, I implemented Grad-CAM explainability, which gave insights on the model behav

- One of the few candidates selected for the developer course, learning various courses related to deep learning specialization through Coursera. -Applying the theoretical knowledge by building models using NumPy, and Tensorflow.

1. Researched and implemented advanced deep neural networks for instance segmentation, semantic segmentation, and object detection using Python and TensorFlow. 2. Investigated Bayesian active learning techniques to enhance data efficiency. 3. Authored a comprehensive knowledge base for ML engineers, documenting models and optimization strategies.

1. Conducted bi-weekly tutorial on topics related to Linear Algebra, ML algorithms, Computer Vision and Robotics for 100+ students. 2. Collaborated with other TAs to design assignments and grade examinations.

1. Conducted weekly tutorials for 100+ students on basic to advanced concepts of Computer Science. 2. Collaborated with professor to grade examinations and assignments of all BSc. Computer Science Students.