Vishal Verma

About Me

I'm a Machine Learning Engineer with extensive experience in causal inference, distributed systems, and ML at scale. Currently pursuing my Master's in Electrical and Computer Engineering at Carnegie Mellon University.

My expertise lies in developing scalable machine learning solutions that solve real-world problems. I've successfully built causal inference platforms, fraud detection systems, and implemented innovative approaches to customer analytics and engagement.

I'm passionate about pushing the boundaries of ML technologies and contributing to the open-source community. My recent work includes scaling causal algorithms for industrial applications and developing fraud detection frameworks for fantasy sports platforms.

Get In Touch

Skills & Expertise

Languages

Python
Java
SQL
Scala
R

Frameworks & Tools

Apache Spark
Scikit-Learn
Ray
TensorFlow
PyTorch
Docker
Flask, Gradio
Pandas

Cloud

Amazon Web Services
Databricks
Azure

Areas of Interest

Causal Inference
Machine Learning
Distributed Systems
AB Testing and Experimentation
MLOps
Deep Learning
NoSQL
Generative AI
AI Safety and Alignment
Mechanistic Interpretability

Work Experience

Sep 2019 – Jan 2025

Lead Machine Learning Engineer

Dream11, Mumbai, India

Achieved 150% faster runtime by designing and building scalable causal inference platform by scaling existing algorithms and causal inference techniques in a distributed manner
Implemented uplift modeling for preventing churn and promotion ranking
Achieved 200% reduced runtime identifying root cause by implementing the RCA detection algorithm using causal discovery and causal algorithms
Saved 60% of resources under-utilization by building a concurrency prediction model using LSTM to ensure app services are future-proofed based on anticipated user demand
Designed and built Scalable Real-time and Batch Fraud Detection ML Systems using a connected component algorithm and graph database bringing down detection and blocking time up to 150%

Dec 2017 – Mar 2019

Machine Learning Engineer

Kohl's (consultant), San Jose, CA

Collaborated with Kohl's data science team to develop a GCP-based Data Science Framework utilizing machine learning models (e.g., Logistic Regression, Random Forest) to analyze customer affinity towards product signals
Impacted over 50 million U.S. customers for enhanced recommendations, discount optimization, and improved search experience

June 2016 – Sep 2019

Sr. Data Engineer

Exadatum Software Services Pvt. Ltd, Pune, India

Developed a versatile, self-serve tool over Apache Spark for solving batch, streaming, and machine learning use cases, with Data-Visualization support
Tailored solutions for use by Data Engineers, Data Scientists, and Business Analysts
Built Data Quality Framework involving ML use-cases such as anomaly detection and trend detection

Projects

Scalable Causal Inference Platform

Designed and implemented a distributed causal inference platform that scales existing algorithms for industrial-scale data.

Python Ray Causal Inference

FENCE: Real-Time Fraud Detection

Fairplay Ensuring Network Chain Entity for Real-Time Multiple ID Detection at Scale in Fantasy Sports.

Graph Algorithms ML Real-time

Concurrency Prediction with LSTM

Developed a concurrency prediction model using LSTM to optimize resource allocation based on anticipated user demand.

Deep Learning LSTM Resource Optimization

Publications

FENCE: Fairplay Ensuring Network Chain Entity for Real-Time Multiple ID Detection at Scale In Fantasy Sports

Upreti, Akriti; Verma, Vishal; Kothari, Kartavya; Thukral, Utkarsh

ACM AIMLSystems'23

Read Paper

Accelerating Causal Algorithms for Industrial-scale Data: A Distributed Computing Approach with Ray Framework

Verma, Vishal; Reddy, Vinod; Ravi Jaiprakash