I'm a passionate AI/ML researcher and developer with a unique background in Chemistry from IIT (BHU) Varanasi, bridging the worlds of molecular science and artificial intelligence. I am also deeply interested in leveraging generative AI—such as GANs and autoencoders—to push the frontiers in material science and environmental science.
| Domain | Technologies & Expertise |
|---|---|
| Programming Languages | Python, C++, C |
| Machine Learning & Deep Learning | PyTorch, Scikit-Learn, Transfer Learning, Model Optimization |
| Computer Vision | Object Detection, Segmentation, YOLO, MobileNet, etc |
| Natural Language Processing | BERT, LLMs, LangGraph, RAG |
| Generative AI | GANs, Autoencoders, and Variants—exploring applications in materials discovery and environmental remediation |
| Development Tools | PyTorch, OpenCV, Hugging Face, Roboflow |
- Graph Neural Networks (GNNs): For molecular structure analysis and drug discovery applications.
- Vision Transformers (ViTs): For high energy physics data analysis, particularly for CMS jet data classification at CERN.
- Advanced Network Pruning & Optimization: To create efficient models for edge devices.
- Real-Time Image Dehazing with Attention Mechanisms: Improving computer vision under challenging environmental conditions.
- LangGraph and RAG Systems: Building domain-specific knowledge agents for scientific applications.
- Hydrological Modeling with Deep Learning: For water quality prediction and resource management.
- Computational Chemistry Tools Integrated with AI: Accelerating materials discovery.
- Multi-modal Learning Frameworks: Combining chemical structure data with visual and textual information.
- Generative AI in Material Science & Environmental Science: Experimenting with GANs and autoencoders to simulate novel materials and predict environmental impacts, offering insights into sustainable innovations.
- Student Wing Member (Project Intern) at Smart Laboratory on Clean Rivers (SLCR)
- Autonomy Team Member at Team AVERERA, developing ML/DL models for autonomous campus shuttles
- Junior Member in ROBOREG (Robotics Research Group)
- Road Segmentation using MobileNetV3: Achieved 0.9 IOU with an optimized architecture for autonomous navigation.
- Network Pruning of ResNet-50: Reduced parameters by 50% while improving accuracy (using ADMM).
- LangGraph-Based Insurance Agent: Integrated speech and LLMs for insurance policy assistance.
- Fake News Classification Using BERT: Achieved 99.81% accuracy with a transformer-based approach.
- Emotion-Emulating Visual AI Chatbot: Created a visually interactive emotional chatbot.
- Real-Time Image Dehazing Using cGAN: Developing attention-based dehazing for robotics and autonomous vehicles.
- Research projects at the intersection of chemistry and AI.
- Applications of computer vision and generative models in various domains.
- Dataset development for specialized applications.
- Environmental monitoring, water quality analysis, and sustainable AI solutions.
- Pioneering work in applying generative AI to material science and environmental science.
- Autonomous navigation systems and intelligent transportation solutions.
- Email: [email protected]
- LinkedIn: Gaurav Singh
- GitHub: GauravSRC