Primary Goals

Our goal is to accelerate scientific discovery by transitioning from human-operated equipment to fully autonomous, closed-loop systems. We achieve this by reframing complex materials research into reward-driven optimization problems – shifting from traditional, myopic measurements to non-myopic active learning that intelligently targets the most valuable data.

To orchestrate this, we pioneer Agentic Models: distributing intelligence across specialized LLM agents that autonomously handle hypothesis generation, protocol translation, and real-time hardware execution. By seamlessly unifying high-throughput synthesis with advanced characterization (SPM, STEM, and automated nanoindentation), we aim to uncover hidden structure-property relationships, manipulate matter atom-by-atom, and permanently close the materials discovery loop.

Machine Learning for Automated Experiment

Agentic Workflow

Bridging human intent and laboratory hardware. We utilize LLMs as cognitive agents to parse complex user goals, seamlessly orchestrate machine-specific instruments, and autonomously run experiments to generate targeted scientific discoveries.

Optimization Workflows

Developing a robust suite of optimization workflows for instrument tuning and materials discovery. By transitioning to non-myopic, reward-driven strategies, we autonomously explore vast combinatorial and random libraries.

News & Events

TBA: Open Data

Future guest lectures, event dates, and group updates will be posted here!

Automated High Throughput & Combinatorial Synthesis

AHMADI Research Lab

Accelerating Hybrid Materials by Automated Discovery Innovation.

Pioneering AI-driven automated synthesis and high-throughput workflows to solve global renewable energy and optoelectronic challenges.

Center for Materials Processing

Directed by Philip Rack

Fostering interdisciplinary research in materials processing to drive new product development, solve performance challenges, and improve industrial competitiveness.

Random Libraries

Exploring vast material spaces through the automated generation and characterization of random libraries.

Techniques & Workflows

AE SPM

Pioneering intelligent SPM workflows to unravel complex material behaviors, accelerating the pace of discovery and enabling transformative advances across diverse fields.

Read More

AE STEM

Developing cutting-edge automated STEM methodologies that leverage machine learning algorithms and autonomous agents for real-time decision-making and adaptive experimentation.

Read More

AE Nanoindentation

Integrating machine learning and adaptive alignment strategies to automate nanoindentation, enabling both targeted microstructural analysis and high-throughput screening of complex combinatorial libraries.

Read More

Xray

Currently in development.

Future video and content will go here.