B.Tech CSE (AI & ML) graduate with internship experience at NIT Warangal & DRDO. Skilled in Python, PyTorch, scikit-learn & LightGBM, building AI solutions in forecasting, classification & computer vision.

Remote
Delhi, India







Bachelor of Technology - BTech, Computer Science AI ML
Maharshi Dayanand University (MDU), Rohtak
12th, Non medical
DAV Multipurpose Public School - India
Join DEEPANSHU and 10k+ other professionals on
Tal is the app where bosses meet engineers & builders directly, without middlemen
Download Tal BossB.Tech CSE (AI & ML) graduate with internship experience at NIT Warangal & DRDO. Skilled in Python, PyTorch, scikit-learn & LightGBM, building AI solutions in forecasting, classification & computer vision.
B.Tech CSE (AI & ML) graduate with research and ML internship experience at NIT Warangal & DRDO. Skilled in Python, PyTorch, scikit-learn & LightGBM, with hands-on experience building real-world AI/ML solutions.
Sonipat
claude code
codex
hugging face
langchain
huggingface
Napkin AI
Research Intern at National Institute of Technology Warangal, Aug 2025 - Dec 2025 · 5 mos, Remote, Built an end-to-end ML pipeline for crop-yield forecasting across 572 Indian districts, leveraging large-scale geospatial datasets from NASA POWER, MODIS, and SoilGrids. Developed and optimized a LightGBM model achieving R² = 0.906 and MAE = 17.65 kg/ha, with training approximately 40× faster than a deep neural network. Performed feature engineering, feature-importance analysis using SHAP, and error profiling to improve model interpretability and performance. Worked with Python, PyTorch, scikit-learn, and LightGBM to build scalable and production-oriented ML solutions.
Machine Learning Intern at INMAS, Defence Research and Development Organization (DRDO), Jun 2025 - Aug 2025 · 3 mos, Delhi, India, Developed and evaluated six machine learning classifiers for toxic-gas identification using multi-sensor chemical data. Achieved 95.02% classification accuracy and an exceptional ROC-AUC of 0.9965 using XGBoost. Performed model evaluation and comparison to identify the best-performing classifier for reliable gas detection. Exported trained models in PKL and ONNX formats for potential integration into real-time drone and embedded-system applications.
Bachelor of Technology - BTech, Computer Science AI ML, Maharshi Dayanand University (MDU), Rohtak
12th, Non medical, DAV Multipurpose Public School - India
github: https://github.com/I-Deepanshu
medium: https://medium.com/@deepanshusnpt