Hi, I'm Nikhil Manakkal
Rising Senior at Lehigh University | ML Engineering, Full Stack & Meteorology
"Rising Senior at Lehigh University with strong domain knowledge in natural science and experience (project & internships) with scientific field work, full stack development, and ML engineering. Passionate about the intersection of computers, machine learning, and meteorology. Proven capability to cross traditional disciplinary boundaries as a researcher."
Core Expertise & Domains
Work Experience
Hands-on internships and research positions across atmospheric science, oceanography, full-stack development, and ML engineering.
William M. Lapenta Internship
Overview
Developed MCP services and chat application for plain language interaction with Numerical Weather Prediction (NWP) data, enabling questions about the weather, spatial coverage mapping, and AI-powered briefing generation through a chat interface.
Key Achievements & Responsibilities
- •Built Model Context Protocol (MCP) services linking AI models directly to NWP outputs from the National Blend of Models (NBM) and Rapid-Refresh Forecast System (RRFS).
- •Designed a chat interface enabling operational meteorologists to query complex spatial weather datasets in natural language.
- •Automated spatial coverage mapping through chat interface and AI-assisted briefing generation for critical weather events.
Project Demo Video
NOAA Presentation Deck
Visual Gallery
Machine Learning Engineer
Overview
Co-developed machine learning pipeline to predict personality type based on Big Five (OCEAN) personality traits from interview transcripts with SHAP explainability for most important features.
Key Achievements & Responsibilities
- •Constructed NLP and machine learning pipelines extracting Big Five personality traits from interview transcripts.
- •Implemented SHAP (SHapley Additive exPlanations) values for model interpretability and feature importance analysis.
- •Evaluated model performance and output insights across varied massive Reddit dataset, achieving macro F1-score of 0.67, comparable to literature benchmarks.
Visual Gallery
Undergraduate Research Assistant
Overview
Paid summer research position performing UV-Vis spectroscopy on sea water samples to identify hydrothermal plumes in vertical profile of water samples collected from scientific cruise. Participated in 3-week long cruise aboard R/V Atlantis to better understand hydrothermal vent systems associated with the Juan De Fuca (JdF) ridge off the coast of the northwestern US and Canada. Trained in QGIS for mapping critical information for the science party aboard the vessel. Performed additional laboratory work on sediment samples from cruise including sectioning, and freezedrying sediments from sea floor cores.
Key Achievements & Responsibilities
- •Conducted UV-Vis spectroscopy to identify chemical hydrothermal plume signals in deep-sea water columns.
- •Participated in a 3-week scientific oceanographic cruise aboard R/V Atlantis on the Juan De Fuca Ridge.
- •Created GIS spatial maps in QGIS for real-time vessel operations and science party navigation.
- •Processed seafloor sediment cores through precision sectioning and freeze drying laboratory protocols.
Visual Gallery
Supervised Research Internship
Overview
Developed automated R-scripts for cleaning, grouping, and performing statistical T-tests on particulate matter (PM2.5) air quality data from multiple sensor locations throughout the Lehigh Valley.
Key Achievements & Responsibilities
- •Engineered automated R analytics pipeline for hourly PM2.5 air quality sensor data.
- •Performed statistical hypothesis testing (T-tests) comparing particulate concentration profiles across sensor sites.
- •Generated meteorological wind rose visualizations linking wind vectors with PM2.5 dispersion.
Visual Gallery
Featured Projects
Key software applications, start-up products, and AI vision tools built from scratch.
CrescentMap
Project Overview
Start-up spatial campus event locator and community builder app. Maintains an interactive map of campus, organizes event chats, 1-mile community chat, direct messaging of friends, and dynamic user profiles.
App Screenshots & Visuals
HistoricalTwin
Project Overview
Find your historical twin and destiny in the Museum of Prague with the help of AI-powered facial similarity matching.
App Screenshots & Visuals
Let's Connect
Whether you're interested in research collaboration, ML development, or full-stack opportunities, feel free to reach out.
Email Address
Direct contact for inquiries & project discussions
LinkedIn Profile
Professional network, research updates & career history