Predicting the impact of pollution at every location on the planet

A team of deep tech entrepreneurs using machine learning to predict hyper-local air pollution prediction

Who We Are

We are an award-winning team of computer vision researchers from MIT and Argonne National Research Lab.

We are building the first open source engine to predict the impact of air pollution at every location on the planet.

KEY CHALLENGES

Air pollution costs businesses $8.1 trillion annually.

Health

6 % of GDP lost due to mortality and morbidity from air pollution. We provide hyper-local air pollution exposure metrics to calculate the health burden of disease caused from pollution exposure.

Operational Efficiency

Air pollution reduces the shelf life of IT assets costing businesses $6.7 billion in energy costs and replacement costs. We perform site-suitability analyses and model improvements in HVAC efficiency over existing baselines while maintaining safety and operational constraints.

Agriculture.

Air pollution causes 10% loss in agricultural yields causing losses across the agricultural and food value chain. We provide air pollution impact assessments for farms to intelligently manage their farming practices.

Regional Partnerships

Projects

Percent of Energy Efficiency Savings

Million children's exposure mapped

Features

Your ML-powered climate sensor stack.

Global Pollution API

Adaptive AI via ML orchestration

General Sensor Intelligence framework

Impact prediction to replace expensive simulation

Few-shot prediction to replace expensive simulation

Geo-visualization engine

Our hierarchical machine-learning model works by integrating sensor and satellite information to get a context-aware picture of pollution over your facilities.

General sensor intelligence framework

A custom-tuned model for one system may not be applicable to another. Therefore, a general intelligence framework is needed to understand the network's interactions.

Few-shot prediction to replace expensive simulation

How we operate equipment and the environment interact with each other in complex, nonlinear ways. Traditional formula-based engineering and human intuition often do not capture these interactions.

Adaptive AI via ML orchestration Systems

Systems cannot adapt quickly to internal or external changes (like the weather). This is because we cannot come up with rules and heuristics for every operating scenario.

Team

Our hard working team

Prithviraj Pramanik

Chief Executive Officer

Prithviraj Pramanik is the CEO of AQAI and is a serial entrepreneur, Ph.D. Candidate and a Fulbright Fellow who has studied cost-effective urban air quality measurement techniques extensively. Mr. Pramanik has worked on one of the dense real-time air quality sensor networks in the US deployed across Chicago.

Christina Last

Cheif Technical Officer

Christina Last is the CTO of AQAI and MIT graduate student, where she holds the 2022-23 US-UK Fulbright All disciplines award. Before MIT, Christina worked at the Alan Turing Institute as a Research Scientist, and as a Data Science Fellow in the Machine Learning Department at Carnegie Mellon University.

Vipul Siddharth

Open Source Technical Mentor

Helping humanitarian innovators understand and leverage the power of Open Source. Previous community manager at Redhat and Fedora.

Madison Marks

Business Advisor

Madison is currently working as a portfolio strategist with the UNICEF Innovation Fund. Previously, Madison was Managing Director of the Social Innovation Lab with Johns Hopkins University.

Our Partners

Fighting pollution together.

Blog

Media Coverage

AQAI technology used to show Cancer rates in Northern Iraq double in affected provinces after facilities resume production in recent years.

Read More

Team wins OpenAI Climate Hackathon

Read More

Christina Last Named a AGU Grand Prize Winner for Data Visualization

Read More

Team becomes 1/9 DS and AI ventures Funded by UNICEF Office for Innivation

Read More
Designed by BootstrapMade