ASHVIN ROKADE
MSc (Research) · Hamilton, New Zealand
AI research · full-stack development
🇳🇿AVAILABLE IN HAMILTON, NZ
Student Visa · No work-hour restrictions · Valid Jul 2027
Seeking junior · grad · intern roles in NZ
ABOUT ME
MSc (Research) student at the University of Waikato. Completed a dissertation on explainable AI for clinical triage; now researching AI-assisted interpretation of milk culture test plates. Previously a full-stack developer in fintech and an independent product builder. Seeking junior, graduate, or intern software roles in New Zealand.
Current Education
HAMILTON, NZ · FEB 2026MSc (Research), Computer Science — University of Waikato
Current research (COMPX594): AI-assisted interpretation of milk culture test plates using visual recognition and expert decision rules.
A+ Deep Learning (COMPX525) · A+ Parallel Programming (COMPX553)
Completed dissertation (COMPX591, A-): XAI for Clinical Triage Prediction using MIMIC-IV
B.E., Computer Science — IET DAVV, Indore · Aug 2016 — Oct 2020 · CGPA 7.89/10
Background
Three years at Nuclei (fintech, FedMobile) · four independent production apps including CrickSnap (786 users) · co-founded Groky. Student visa with no work-hour restrictions; available immediately.
PROFESSIONAL WORK
TECHNICAL EXPERTISE
React • Next.js • TypeScript • JavaScript • Android • HTML/CSS • TailwindCSS
Node.js • Java • Spring Boot • Python • RESTful APIs • Microservices • WebSocket
MongoDB • PostgreSQL • MySQL • Redis • Elasticsearch
AWS • Docker • Kubernetes • OpenShift • Git • CI/CD • ELK Stack (Elasticsearch, Kibana)
Kafka • Redis Pub/Sub • Message Queues
Python • XGBoost • SHAP • scikit-learn • Streamlit • Computer Vision • pandas • NumPy • LLMs
MSc Research
UNIVERSITY OF WAIKATO · HAMILTON, NZ
- •Applying visual recognition and expert decision rules to interpret milk culture test plates
- •Trained XGBoost model on MIMIC-IV ICU dataset achieving 96.1% AUC-ROC
- •Built explainable AI interface using SHAP for clinical audiences
Coursework: A+ Deep Learning (COMPX525) · A+ Extremely Parallel Programming (COMPX553)
NUCLEI
FINTECH INFRASTRUCTURE PLATFORM · BENGALURU, INDIA
Software Development Intern
Full-time placement delayed to Feb 2021 due to Covid-19
Developed SDK modules for travel booking services deployed across 7+ major Indian banking applications
- •Built and deployed Bus, Flight, and Hotel Booking Modules for Nuclei SDK platform
- •Implemented scalable microservices architecture serving multiple banking partners
Full Stack Developer (FedMobile)
Led full-stack development and infrastructure modernization for Federal Bank's mobile banking platform
Frontend Development
- •Developed features and resolved critical bugs for FedMobile application
- •Enhanced user experience for Federal Bank's primary mobile banking platform
Backend & Middleware
- •Built middleware layer connecting bank servers with frontend applications
- •Designed RESTful APIs handling secure banking transactions
DevOps & Infrastructure
- •Architected logging infrastructure using ELK stack for production monitoring
- •Led migration of backend applications from IBM Cloud to OpenShift platform
INDEPENDENT PROJECTS & BUILDING
MAY 2023 — JAN 2026 · SHIPPED PRODUCTS · REAL USERS
AI Developer Tooling
- •Developed AI-based code analysis tool for developers
Product Development & Launch
- •Developed and launched Chinese learning platform
- •Built comprehensive interior design platform
- •Built online multiplayer card game platform with real-time gameplay
- •Built full-stack cricket scoring application serving 786 active users
- •Built for a 13-day, 64-team knockout tournament I organised — scoring, fixtures, and live updates
FEATURED WORK
Projects
Production applications and research software with real-world impact
Milk Culture Plate AI
MSc Research — COMPX594 · In Progress
Current MSc research at the University of Waikato. Developing an AI-assisted system to interpret milk culture test plates using visual recognition and expert decision rules — combining computer vision with domain knowledge for veterinary diagnostics.
XAI Triage
Hospital Mortality Risk — AI Explainability Demo
Completed MSc dissertation (COMPX591, University of Waikato, A-): XAI for Clinical Triage Prediction — A Benchmark Study Using MIMIC-IV. Predicts ICU patient mortality risk using a trained XGBoost model and explains every prediction in plain language using SHAP (Shapley values). Trained on MIMIC-IV — real ICU data from Beth Israel Deaconess Medical Center.
CrickSnap
Cricket Scoring Application
A full-stack cricket scoring application built with Next.js 16, TypeScript, and Supabase. Features role-based access control, live ball-by-ball scoring, real-time updates, and comprehensive match management. Currently used by cricket scorers across India.
Let's Build Something Great Together
🇳🇿Open to Opportunities in New Zealand
Available in Hamilton, NZ • Open to remote opportunities