Boston, MA | nikhilanilprakash@gmail.com

I’m a Master’s student in Electrical & Computer Engineering at Northeastern University, specializing in Machine Learning, Computer Vision, and Robotics Systems.
I build end-to-end ML systems, real-time robotics pipelines, and production-grade AI tools that combine mathematical rigor with practical engineering.

I’m currently seeking Full-Time roles (2025) in:
Machine Learning • Computer Vision • Robotics • MLOps • Software Engineering

Machine Learning, Bayesian Modeling, Time-Series Forecasting
View Project

• Engineered 4+ models (Bayesian Regression, Gaussian Process, Bayesian Neural Nets, Decision Trees) over 7,000+ NOAA records.
• Decision Trees achieved 37.9% RMSE reduction, R² = 0.41, and best MAE among all models.
• Built probabilistic models with Laplace Bayesian MLP, achieving 92–95% CI coverage (best uncertainty calibration).
• Led feature engineering with lagged variables, seasonal encoding, and rolling window statistics.

Sports Analytics, Regression Modeling
View Project

• Built predictive models (Linear Regression, XGBoost, AdaBoost, Decision Trees) on 500+ IPL matches across 15 seasons.
• Achieved 82% accuracy, with a 15% uplift from custom feature engineering and preprocessing.
• Delivered complete evaluation pipeline with error analysis, learning curves, and match-level score forecasts.

Computer Vision, 3D Geometry, Feature Matching, Optimization
View Project

• Enhanced image clarity with CLAHE and extracted SIFT features for multi-view reconstruction.
• Estimated camera poses and refined structure via bundle adjustment in GTSAM.
• Achieved 51.77% improvement in reconstruction accuracy using Levenberg–Marquardt optimization.
• Visualized optimized 3D point clouds using Matplotlib and Python.

Robotics, Sensor Fusion, Pose Estimation, ROS2
View Project

• Implemented LiDAR–IMU fusion using pre-integration, factor graph optimization, and SIFT-based frame matching.
• Evaluated on KITTI with ATE/RPE metrics; visualized trajectories in RViz.
• Integrated the pipeline as ROS2 nodes and simulated SLAM in Gazebo environments.

3D Mapping, SLAM, Factor Graphs
View Project

• Benchmarked both SLAM systems on KITTI and Gazebo trajectories.
• LeGO-LOAM: ATE = 65.24 m, RPE = 0.64 m
• LIO-SAM: ATE = 248.17 m, RPE = 4.24 m (significant IMU drift).
• Conducted full error analysis, algorithmic comparison, and shape-preservation tests in indoor simulation.
• Produced actionable insights on robustness, drift behavior, and sensor dependency.

Control Systems, Stability Analysis, Real-Time Systems
View Project

• Proposed two novel control schemes—PSET & PCMSET—combining periodic sampling and switched ETC.
• Reduced event transmissions by 30–40% compared to existing CET/PET/SET/MSET methods.
• Verified stability on second-order and fourth-order inverted pendulum systems.
• Identified optimal operating periods (IOP): p = 0.6 for PSET, p = 0.5 for PCMSET.
• Included full simulation results with transmission counts, convergence plots, and stability analysis.

• Implemented an end-to-end SEO & performance monitoring system using Python ETL + anomaly detection.
• Combined rule-based + ML models to compute health scores across 100+ webpages.
• Built automated BI reporting with Markdown and dashboards for web vitals and metadata compliance.

• Developed inverter software in ASCET for electric vehicles; built a safety-state switching module.
• Tested control modules on a remote simulator; improved system reliability and documentation quality.

• Designed a speed-monitoring algorithm improving calibration precision by 15%.
• Built a Python-based statistical dashboard reducing calibration time by 30%.
• Standardized software components, reducing warnings by 25%.

Programming: Python, C, C++, R, SQL, SystemVerilog, JavaScript
ML / CV: NumPy, Pandas, TensorFlow, OpenCV, Scikit-learn, Bayesian Models, GTSAM, Open3D
DevOps / MLOps: Docker, Airflow, DVC, GitHub Actions, Linux, GCP
Robotics: ROS/ROS2, Gazebo, LiDAR SLAM, Sensor Fusion, State Estimation
Data / Analytics Tools: Tableau, ETL Pipelines, Statistical Modeling
Certifications: Google Data Analytics, IBM Data Science, DeepLearning.AI ML Specialization

• Email: nikhilanilprakash@gmail.com
• LinkedIn: Add your link
• Portfolio: Add your domain (e.g., nikhilaprakash.github.io)

Thank you for visiting!
Feel free to explore my projects or reach out if you'd like to collaborate.