Sharan Ramjee

I’m a Machine Learning Engineer passionate about bridging the gap between AI research and production systems. I hold a Master’s degree in Computer Science (AI concentration) from Stanford University and a Bachelor’s degree in Computer Engineering from Purdue University.

When I’m not training models, you’ll find me inventing new cocktails, DJing house and techno, chasing waves surfing, or traveling to exotic destinations. If you’d like to chat about my work or anything else, feel free to reach out to me at sharanramjee [at] gmail [dot] com.

LinkedIn GitHub Scholar
Resume CV

Industry Experience

Stripe

2022 - Present

Machine Learning Engineer - Fraud Discovery

  • Led Project Scorpion - first time-series transformer for fraud detection
  • Saves $8M/year in fraud losses
  • Built real-time ML inference pipeline serving millions of requests

Google

Summer 2019

Software Engineering Intern - Cloud AI

  • Worked on model distillation for production ML systems
  • Contributed to TensorFlow Extended (TFX) pipeline

Palo Alto Networks

Summer 2021

Software Engineering Intern - Cortex XDR

  • Built ML-powered threat detection features
  • Project Aerial: CAASM integration

Qualcomm

Summer 2018

Machine Learning Intern - QoS Optimization

  • Developed ML models for Quality of Service optimization

Research

Natural Language Processing

LLMs Sentiment Analysis Cyberbullying Detection

Computer Vision

Action Recognition Depth Estimation Super-Resolution

Signal Processing

Wireless Communications Modulation Classification Interference Detection

Graph Learning

GNNs Graph Embeddings Network Analysis

Publications

Ensemble Wrapper Subsampling for Deep Modulation Classification

IEEE TCCN (2021)

View Paper

Efficient Training of Deep Classifiers for Wireless Source Identification

IEEE WCL (2020)

View Paper

Deep Learning for Interference Identification: Band, Training SNR, and Sample Selection

IEEE SPAWC (2019)

View Paper

Fast Deep Learning for Automatic Modulation Classification

IEEE MLCETI (2019)

View Paper

Projects

NLP
Transformers
Cyberbullying Detection with Attention-Weighted IG
Aspect-Target Sentiment Classification
Computer Vision
CNNs & GANs
Single-Image Stereo Depth Estimation
Skeleton-based Action Recognition
Pothole Detection with Super-Resolution
Graph Learning
GNNs
Gradient Boosting with Wasserstein Embeddings
Meta Learning
Few-Shot
FLITE: Memory-Efficient Meta Learning
ML Systems
Production ML
AutoChef: Ingredient-to-Recipe CV

Teaching

Head Teaching Assistant

Stanford University

Course Assistant

Stanford University

Course Assistant

Stanford University

Leadership

Purdue IEEE Computer Society

President

2018 - 2019

TreeHacks

Organizer

2021 - 2022

Autonomous Motorsports Purdue

Software Lead

2017 - 2019

Undergraduate Research Student Program

Research Assistant

2017 - 2020

Talks

Model Distillation

Google Cloud AI

2019-12-04

QoS Optimization with ML

Qualcomm

2019-08-15

Deep Learning for Interference Identification

IEEE SPAWC

2019-07-02

Deep Learning for Style Transfer

Purdue IEEE CSOCIETY

2018-11-15

Skills

Languages

Python
Java
C++
JavaScript
SQL

ML Frameworks

PyTorch
TensorFlow
Hugging Face
JAX
scikit-learn

ML Domains

NLP / LLMs
Computer Vision
Graph Neural Networks
Reinforcement Learning
Time Series

Infrastructure

AWS
GCP
Docker
Kubernetes
Spark