Available for Opportunities

Josh Jestine

AI Engineer & ML Specialist

Building ML systems that work in practice: From predictive maintenance pipelines to reinforcement learning agents that beat human baselines.

72%Prediction Accuracy
83.9%Efficiency Gain
1M+Training Episodes
56→9Steps per Task
200+Developers Mentored
40+IoT Sensors
15+Projects Shipped
72%Prediction Accuracy
83.9%Efficiency Gain
1M+Training Episodes
56→9Steps per Task
200+Developers Mentored
40+IoT Sensors
15+Projects Shipped
Scroll

Engineering ML Systems That Actually Ship

I'm an AI/ML engineer with a Master's in Artificial Intelligence from Northeastern University, with a background in Information Technology and Cybersecurity from University of Mumbai.

My work ranges from building predictive maintenance models that cut elevator downtime by 26%, to training deep RL agents that reduce task completion from 56 steps down to 9. I like working on problems where the numbers actually move.

3+
Years of Experience
15+
Projects Completed
200+
Developers Mentored

Academic Foundation

A strong theoretical foundation combined with hands-on research experience

Master of Science in Artificial Intelligence

Northeastern University, Boston
Dean's List Merit Scholarship recipient. Focused on advanced algorithms, unsupervised machine learning, and AI for human-computer interaction.

Coursework: Algorithms, Unsupervised ML, Machine Learning, PDP, AI for HCI, NLP & FAI
Expected April 2026

Bachelor of Engineering in IT

University of Mumbai
Graduated with Honors in Cybersecurity. GPA: 3.4/4.0. Built a strong foundation in systems, security, and software development.
2020 - 2024

Professional Journey

Industry internships and technical community leadership

Technical Intern

Express Elevators
May 2025 - August 2025
  • Built predictive maintenance ML model achieving 72% accuracy in predicting component failures across 35 elevators
  • Automated fault detection using OpenCV, identifying 8 recurring fault patterns and saving 12 hours/week of manual log review
  • Reduced average wait time by 12% through optimized dispatching algorithms across 3 commercial buildings
  • Built data collection pipeline with Python and MySQL to aggregate sensor data from 40+ IoT devices
  • Decreased elevator downtime from 4.2 to 3.1 hours per incident through usage analytics and trend reporting

Co-founder

Google Developer Student Club | GDSC SFIT
August 2023 - August 2024
  • Directed national-level hackathon with 200+ participants from 15+ universities
  • Managed $10,000 budget for ML, Cybersecurity, and Cloud Infrastructure tracks
  • Led team of 8 to deliver 5+ technical workshops driving 40% growth in engagement
  • Designed and delivered a hands-on Deep Learning workshop using PyTorch for 60+ students, covering image classification with CNNs

Featured Work

Selected projects across ML, computer vision, and data engineering

F1 Race Winner Predictor

Python LightGBM FastF1 Scikit-learn

Lightweight F1 race winner prediction model using FastF1 and LightGBM. Trains on historical race data to predict win probabilities based on qualifying results, driver form, and team performance.

Features rolling averages, circuit-specific modeling, and team performance metrics for accurate race outcome predictions.

70-80%
Podium Accuracy
50-60%
Winner Accuracy
~3 min
Training Time
View on GitHub

SpendLens

Python Streamlit Google Gemini API Pandas

A privacy-focused personal finance app that turns raw transaction data into mood-adaptive "Money Narratives" — spending insights without storing any sensitive user data.

Includes an LLM-powered assistant ("Finn") via the Gemini API for contextual follow-up queries on anonymized stats. Built following Shneiderman's Eight Golden Rules for user control and transparency.

100%
Privacy-First
5
Mood Modes
HCAI
Principles
View on GitHub

Lung Cancer Detection

Python TensorFlow Keras CNN Streamlit

ML-powered lung cancer detection system using Convolutional Neural Networks to analyze CT scan images. Enables early detection when treatment is most likely to be successful.

Features advanced image processing techniques including edge detection, segmentation, and feature extraction for accurate medical diagnosis assistance.

CNN
Architecture
CT Scan
Image Analysis
Live
Deployment
View on GitHub

Adaptive Pathfinder Agent

Python PyTorch OpenAI Gym CNN-DQN

CNN-DQN agent trained across 1 million episodes in a dynamic simulation with moving obstacles and adaptive rewards. Achieved a 35% boost in navigation efficiency.

Uses experience replay, target networks, and hyperparameter tuning to achieve stable convergence over Q-learning baseline.

83.9%
Efficiency Gain
56→9
Steps Reduced
1M+
Episodes
View on GitHub

Deep Cadence Companion

Python Music21 VAE MIDI

Algorithmic music generation system integrating rule-based harmony modeling with variational autoencoders to replicate classical piano textures.

Features tonal extraction, chord progression modeling, and latent-space regularization inspired by Mozart, Bach, and Beethoven.

93%
Beam Similarity
Real-time
MIDI Synthesis
DAW
Integration
View on GitHub

Technical Expertise

Tools and frameworks I use regularly

PyTorch
TensorFlow
Scikit-learn
Deep Learning
Reinforcement Learning
Computer Vision
NLP
CNNs
OpenAI Gym
OpenCV
PyTorch
TensorFlow
Scikit-learn
Deep Learning
Reinforcement Learning
Computer Vision
NLP
CNNs
OpenAI Gym
OpenCV
Python
NumPy
Pandas
Matplotlib
Seaborn
Docker
AWS
Linux
Git
MongoDB
MySQL
Bash
CI/CD
Python
NumPy
Pandas
Matplotlib
Seaborn
Docker
AWS
Linux
Git
MongoDB
MySQL
Bash
CI/CD

Machine Learning & AI

PyTorch, TensorFlow, Scikit-learn, OpenAI Gym. Reinforcement learning (DQN, Q-Learning), deep learning (CNNs, VAEs, RNNs), NLP, computer vision, and model training with feature engineering.

Systems & Infrastructure

Linux (Ubuntu), Windows Server, macOS. Docker, VirtualBox/VMware, AWS, server configuration. Familiar with Agile methodology, network protocols, and hardware troubleshooting.

Development & Automation

Git, GitHub, Bash, PowerShell, Python scripting. CI/CD pipelines, DevOps workflows. Database work with MongoDB and MySQL. Cloud deployment on AWS.

Let's Talk

Open to full-time roles, research collaborations, or interesting conversations about ML and AI engineering.