Vimal Kumar
Machine Learning Engineer • Production ML • Real-Time Systems • GenAI
Hire this person
Send a job offer directly to this candidate
About
Machine Learning Engineer with hands-on experience delivering production ML systems, GenAI applications, and deep learning pipelines across the full development lifecycle. Completed internships at UpSkill Campus & UCT (LSTM-based RUL prediction, XGBoost silica forecasting) and Unified Mentor (92% accuracy clinical models, ₹60K MAE vehicle pricing). Built real-time fraud detection at sub-200ms latency, an enterprise RAG pipeline achieving MRR=0.833, and a live AI recruiter tool in production.
Strong in Python, FastAPI, Docker, and cloud infrastructure with a consistent track record of measurable impact.
Experience
Data Science & ML Project-Based Intern at UpSkill Campus & Uniconverge Technologies (UCT) (2026-03 – 2026-04)
- Built an end-to-end RUL prediction pipeline on the NASA CMAPSS dataset (FD001–FD004) using a 2-layer LSTM model with Dropout regularization and EarlyStopping, achieving significant RMSE improvement over a Random Forest baseline.
- Engineered rolling window features (mean, std over 10-cycle windows) and applied piecewise linear RUL labeling (cap=125 cycles) to prepare multivariate time-series sensor data for deep learning.
- Developed a silica impurity prediction model for a real industrial flotation plant dataset using XGBoost with GridSearchCV hyperparameter tuning, achieving strong R² scores on both standard and iron-feature-excluded variants.
- Delivered complete project reports and GitHub repositories covering EDA, preprocessing, model design, implementation, results, and learnings as per UCT internship deliverable standards.
Machine Learning Intern at Unified Mentor Private Limited (2025-06 – 2025-07)
- Developed and evaluated ML models on 5,000+ clinical records for liver cirrhosis staging, achieving 92% diagnostic accuracy and supporting data-driven clinical insights.
- Built and validated a vehicle price prediction model achieving ₹60K MAE using XGBoost and Optuna hyperparameter tuning on real-world automotive datasets.
- Improved model performance by 15% through systematic feature engineering, stratified cross-validation, and structured experimentation workflows.
- Established ML best practices across the team including Git-based versioning, reproducible training pipelines, and deployment-ready model packaging.
Education
B.Tech in Computer Science & Engineering – Rajkiya Engineering College (2020-11 – 2024-07)
