Exabyte Inc., the developer of Mat3ra.com, is proud to announce its selection under the U.S. Department of Energy's Genesis Mission as part of three Phase I projects — including AlphaFilm, a Mat3ra-led effort to build a first-in-the-nation closed-loop agentic AI capability for semiconductor thin-film materials.
Exabyte Inc., the developers of Mat3ra.com, is proud to announce its selection under the U.S. Department of Energy’s (DOE) Genesis Mission (DE-FOA-0003612), a national initiative to harness artificial intelligence and the DOE’s world-leading scientific infrastructure to accelerate discovery across materials, energy, and manufacturing. Mat3ra.com will participate in three selected Phase I projects, spanning AI-driven materials processing, quantum materials discovery, and the cybersecurity of scientific AI workflows.
Mat3ra.com is especially excited to lead AlphaFilm, a first-in-the-nation closed-loop agentic AI capability for semiconductor thin-film materials, selected under the Genesis Mission's challenge area on Reenvisioning Advanced Manufacturing and Industrial Productivity (AI-Driven Materials Processing). AlphaFilm unites machine-learned interatomic potentials and agentic AI orchestration on the Mat3ra.com platform with combinatorial thin-film synthesis and automated multimodal microscopy, forming an autonomous experiment–computation loop that aims to cut materials-screening cycles from weeks to hours for ferroelectric wurtzite nitride films — a transformative platform for non-volatile memory, RF filters, neuromorphic computing, and MEMS.
The project builds directly on Mat3ra.com's recent case study with the group of Prof. Sergei Kalinin at the University of Tennessee, Knoxville, in which machine learning extracted atomic-scale features from electron microscopy and fed them back into high-throughput simulations to refine subsequent experiments. In AlphaFilm, this closed-loop paradigm is scaled to industrially relevant deposition: the frontier semiconductor deposition capabilities of the effort are led by the National Lab of the Rockies, whose team under Andriy Zakutayev contributes world-leading combinatorial sputtering of Al(1-x)Sc(x)N libraries with autonomous process feedback, while the University of Tennessee, Knoxville provides automated scanning-probe and multimodal characterization. The project is carried out with support from industrial partner(s), targeting the translation of the resulting nucleation and growth capabilities toward pilot-scale industrial production.
Mat3ra.com is pleased to join "High-Fidelity AI-Ready Excited-State Materials Datasets and AI Models for Accelerated Discovery," led by Lawrence Berkeley National Laboratory under Prof. Steven G. Louie, together with University of Southern California, Carnegie Mellon University, Yale University, and NVIDIA. Selected under the Genesis Mission's focus area on Functional to Quantum Materials, the project will generate high-throughput, high-fidelity GW and GW-BSE many-body datasets with the exascale-performant BerkeleyGW software and train AI surrogate models for excited-state phenomena in quantum and optoelectronic materials. Mat3ra.com will assist the team with cloud-based high-performance computing, web-based dissemination and collaboration, and the AI-readiness of the data produced.
Mat3ra.com also joins SPOTTER-AI ("Scientific Provenance-Oriented Threat Tracing and Attribution for Genesis Workflows"), led by Argonne National Laboratory under Dr. Bogdan Nicolae, together with Oak Ridge National Laboratory, the Illinois Institute of Technology, and Hewlett Packard Enterprise. Selected under the Genesis Mission's focus area on Data Provenance and Integrity Verification, SPOTTER-AI is an AI-powered cybersecurity framework that uses detailed provenance and importance-based attribution to detect integrity attacks on scientific AI workflows and trace them to their root cause. Mat3ra.com contributes expertise in data curation and domain-specific ontologies, extending its ongoing collaboration with the National Institute of Standards and Technology (NIST) under award 70NANB24H205 on data standards for electronic materials research and development.
“Being selected from among 5,000 submissions from frontier institutions all across the United States is a great sign of respect, and it is both a pride and an obligation for us to be able to contribute to advancing this effort. It highlights the unique position of Mat3ra as a world leader in enabling impactful AI-driven materials science,” said Timur Bazhirov, CEO of Exabyte Inc. “Our recent work at Mat3ra is focusing on enabling AI-native capabilities for the design and development of real-world materials, including thin films, surfaces, interfaces, defects, and other complex effects that govern their behavior. The three projects we are part of are well aligned with our efforts, and we are excited to support the DOE Genesis Mission and contribute to building scientific superintelligence for materials and energy applications.”
The Mat3ra.com platform integrates multiscale physics-based and data-driven materials modeling in the cloud to enable rapid discovery, validation, and optimization of advanced materials, and provides FAIR, AI-ready data infrastructure and workflow orchestration for research teams in industry, academia, and government. Across all three projects, Mat3ra.com serves as a catalyst for American innovation. As the Department of Energy builds the national infrastructure — the “highways and railways” for AI-driven scientific discovery — from leadership-class computing to the American Science Cloud, Mat3ra.com provides the “last-mile delivery” that puts these capabilities directly into the hands of scientists, enabling them to get to results faster, together, in a collaborative and accessible manner.
Team Mat3ra
Exabyte Inc. / Mat3ra.com
press@mat3ra.com
Original Press Release is at this link
Exabyte Inc.
1212 Broadway Plaza, Ste #2100 Walnut Creek, CA 94596
+1 (510) 473 7770
info@mat3ra.com
Mat3ra Japan K.K.
Otemachi Building 6F Inspired Lab,
1-6-1, Otemachi, Chiyoda-ku, Tokyo 100-0004 Japan
