Agnik Sparks Lab

Scaling Intelligence.
Powering Discovery.

Pushing the boundaries of AI through scalable Distributed Machine Learning.

About Us

Agnik Sparks Lab explores new directions and builds on existing ones to solve the next generation of large modeling and control problems. The Lab is backed by extensive expertise, award-winning distributed machine learning research, global innovation, and leadership.

Fortune 500 Clients

Trusted by leading companies for large-scale machine learning deployments

Award-Winning Research

Recognized for breakthrough innovations in distributed computing

30+ Years

Of continuous innovation in distributed machine learning research

Industry Leading

Pioneers in privacy-preserving and scalable machine learning
solutions

Our Research Focus Areas

Pushing the boundaries of distributed machine learning across three critical domains

Vehicle Analytics

Vehicle Analytics

Real-time analytics from vehicle health, driver behavior, fuel usage, and location data in order to optimize fleets and support wide range of connected car services.

Real-Time Bidding

Digital Campaign Management

Deals with automated algorithms for optimizing digital campaigns and computing bids in real-time advertisement-auctions for precise audience-targeting and efficient ad-buys.

Large Language Model

Large Language Model

Developing LLM models using state of the art scalable distributed machine learning algorithms to reshape language understanding and generative AI.

How We Work

Our systematic approach to advancing distributed machine learning research

Model Design &
Theory

We begin with rigorous theoretical foundations, developing novel algorithms and mathematical frameworks for distributed learning systems.

Large-Scale Simulation in Real World

Advanced simulation environments allow us to test our approaches across thousands of virtual nodes before real-world implementation.

Real-World Experimentation

We deploy and validate our systems in production environments, measuring performance, efficiency, and reliability metrics.

Open Evaluation & Deployment

Results are openly published and peer-reviewed, feeding back into our research cycle for continuous improvement.

Each cycle builds upon previous learnings, driving continuous innovation in distributed machine learning

Meet Our Team

Leading experts in distributed machine learning, driving innovation at the intersection of AI and systems research

Dr. Hillol Kargupta

Dr. Hillol Kargupta

Co-Founder, President, & Head of Research

Distributed Machine Learning
Privacy-Preserving Data Mining
Large-Scale Data Science Connected Car Analytics
ML on the Edge for Mobile and IoT applications
Distributed Data Mining Privacy-Preserving Mining Large-Scale Data Science Connected Car Analytics Mobile and Ubiquitous Computing

Dr. Hillol Kargupta is a co-founder and President of the Agnik Group of Companies. He is an IEEE Fellow. Read More

Team Agnik

Agnik Spark Lab is a global team united by innovation, entrepreneurship,
excellence, and a strong commitment to ethics. Here is a picture of some
of the team members from Agnik India.

Media Center

Stay updated with the latest insights, research findings, and industry developments in distributed machine learning

Video thumbnail
United States Vehicle Analytics Market to hit US$ 18.0 Billion by 2030 | Leading Companies - Xevo Inc, Agnik LLC, Microsoft

October 01, 2025

Agnik is a leading force in the rapidly growing U.S. Vehicle Analytics Market, offering AI-powered solutions for predictive maintenance, driver behavior, and fleet optimization. Leveraging big data, IoT, and machine learning, Agnik helps businesses boost efficiency, cut costs, and adapt to the evolving demands of connected, intelligent vehicle ecosystems.

🏆 Conference : ECML- PKDD 2025 Video thumbnail
Organizing Discovery Program in the 2025 European Machine Learning Conference

Predictive Online Digital Sales
(PODS) and Marketing

September 15 - 19, 2025

The Agnik Sparks Lab organized the Discovery Challenge Program on predictive digital campaign management at the 2025 European Machine Learning (ECML/PKDD) conference. Agnik team released real-life campaign performance data and the participants developed scalable ML models to predict Click-Through Rates (CTR) and ad conversions.This was the first public release of Agnik’s campaign management data, supporting research in large-scale online advertising. Agnik Sparks Lab members published a paper in the proceedings of the conference.

Video thumbnail
Top Ten Vehicle Analytics Companies in the World

Vehicle Analytics Market Size, Share & Industry Analysis

September 08, 2025

Agnik is recognized as one of the top 10 key companies in vehicle analytics, empowers the industry with award-winning big data platforms for OEMs, insurers, and fleets. With the global market set to reach $27.73 billion by 2032, Agnik’s cutting-edge real-time analytics drive safer, smarter, and more connected mobility for all stakeholders.

🏆 Conference : ECML-PKDD 2025 Video thumbnail
The Global Hack for Fighting COVID-19

Breathe@Vyncs

April 21, 2020

Agnik was selected as a top-15 winner in The Global Hack for Fighting COVID-19 with its HealShare project. HealShare enables privacy-protected breath audio recordings via smartphones to detect breathing discomfort patterns globally. Agnik’s "Breathe@Vyncs" app feature leverages this technology, advancing safer, data-driven public health and mobility integration. Breathe@Vyncs App Makes COVID-19 Hackathon Finals.