2026 Call for Joint Integrated and Seed Grant Proposals

Abstract

We propose a high-performance electro-optic (EO) modulator combining atomic lift-off (ALO) of single-crystal LiNbO3 with photonic-crystal (PhC) cavity engineering. Using the recently reported ALO process (Nature 2025)6 from the Kim group's layer-transfer family (Nature 2017, 2020, 2023)7–9, a damage-free ultra-thin LiNbO3 (UTFLN, <100 nm) membrane is dry-stamped onto a PhC cavity, placing the active EO material in the strongest optical field. FDTD shows that the <100 nm UTFLN thickness simultaneously enables fast modulation and maximizes the Pockels response — yielding Vπ ≈ 1–3 V at 1550 nm and a modulator footprint <40 μm, the smallest reported for high-speed LiNbO3 modulators. Combining MIT's ALO with INESC-MN's photonic-device fabrication, we will demonstrate a unit-device UTFLN-on-PhC modulator as the building block toward CMOS-voltage modulators for co-packaged optics (CPO) and terabit-per-second optical transceivers in next-generation AI systems.

MIT PI
Jeehwan Kim, Professor, Department of Mechanical Engineering

PT PI
Susana Cardoso de Freitas, Principal Investigator, INESC-MN and Full Professor, Physics Department, University of Lisbon (IST)

Abstract

This project will establish a closed-loop framework for optimizing next-generation 2D semiconductor devices through a collaboration between MIT and the International Iberian Nanotechnology Laboratory (INL). MIT will synthesize high-quality transition metal dichalcogenide (TMD) materials using chemical vapor deposition (CVD) growth and fabricate field-effect transistors and contact structures. INL will perform advanced atomic-scale characterization, using electron microscopy to quantify strain, defects, interfaces, phase purity, local electrostatic fields. The team will use machine learning (ML) techniques to correlate structural characterization with key device metrics such as carrier mobility, threshold voltage, contact resistance, hysteresis, variability, and reliability. The project aims to demonstrate new technologies that improve material uniformity and compatibility with back-end-of-line (BEOL) Si CMOS integration. Expected outcomes include beyond-state-of-the-art 2D devices and a novel an ML-enabled defect–strain–performance correlation pipeline to accelerate lab-to-fab transition.

MIT PI
Tomas Palacios, Professor, Department of Electrical Engineering & Computer Science

PT PI
Paulo Ferreira, Full Professor in the Department of Mechanical Engineering at IST, University of Lisbon; Advanced Electron Microscopy, Imaging and Spectroscopy Center, International Iberian Nanotechnology Laboratory. 

Additional collaborators
Puspendu Guha, PhD, Staff Researcher, INL, Portugal
Pedro Costa, PhD, Staff Researcher, INL, Portugal 
Cristiana Alves, PhD, Facility Manager, INL, Portugal

Abstract

AI coding tools can generate effective “one-shot” solutions, but they struggle with the ongoing evolution of real software products. One reason is structural: when software is not modular, AI systems cannot easily determine where changes belong. Another is a limitation of current AI tools themselves. They rely on ever-growing contexts that include all generated code and prior interactions, which gradually reduces the effectiveness of the underlying language model while greatly increasing token costs.

This project addresses both challenges. It introduces a new structuring method, called concept design, that brings full modularity to software development. It also develops a strategy that enables fine-grained, targeted context instead of passing the entire project history to the model, and verification techniques to ensure correctness of generated code. Together, these techniques make AI coding more reliable, flexible, and cost-efficient, while also providing broader benefits for other forms of agentic AI.

MIT PI
Daniel Jackson, Professor, Department of Electrical Engineering & Computer Science

PT PI
Alcino Cunha, Research Coordinator, INESC TEC, Associate Professor at University of Minho
 

Abstract

Bayesian inference and data assimilation are central tools for building reliable digital twins of physical, environmental, and engineering systems. Yet their reliability often depends on priors whose implications for simulated dynamics are difficult to anticipate. This project develops prior-predictive calibration for dynamical systems: a framework for translating expert knowledge about system behavior into principled priors for Bayesian dynamical models. Rather than specifying priors only over system parameters, we will constrain the trajectories they generate, including stability, boundedness, oscillation periods, extinction or coexistence regimes, conservation properties, and admissible responses. We propose to optimize and learn priors whose prior predictive distributions satisfy these behavioral constraints while remaining close to an initial scientific prior. The project will develop differentiable calibration methods, surrogateassisted admissible-set learning, and decision-oriented evaluation for posterior inference and data assimilation. The MIT–Portugal team combines Bayesian machine learning, uncertainty quantification, mechanistic modeling, and sequential decision-making, delivering open-source software, benchmarks, and a realistic dynamical-system case study.

MIT PI
Youssef Marzouk, Professor, Department of Aeronautics and Astronautics

PT PI
Eliezer de Souza da Silva, Assistant Professor, Department of Informatics Engineering, University of Coimbra

Additional collaborator
Matthew E. Levine, Research Scientist, Basis Research Institute; Iñigo Urteaga, Ramón y Cajal and Ikerbasque Research Fellow, Machine Learning Group, BCAM

Abstract

Climate change is increasing the frequency, duration, and intensity of heat waves across Southern Europe. Portugal’s cities house among the most vulnerable populations due to high building density, aging building stocks, urban heat island effects, and growing cooling needs. Given the high costs of electricity in Portugal, overheating risk and energy poverty are becoming increasingly interconnected, especially in older residential buildings with limited thermal performance. The proposed project builds on the ongoing development of a global building inventory (GloBI), an MIT-led initiative to create climate-actionable information on individual buildings worldwide. In collaboration with the Portugal’s National Institute of Health (INSA) and Energy Agency (ADENE), our longstanding ISTMIT team will create an indoor overheating risk map for every building in the country. The map will enhance Portugal’s national heat warning system ICARO (Silva et al., 2024) and help municipalities to better monitor their constituents’ safety during heat waves.

MIT PI
Christoph Reinhart, Professor, Department of Architecture

PT PI
Dr. Ricardo Gomes, Assistant Professor, IN+ Center for Innovation, Technology and Policy Research; IST-ID, Associação do Instituto Superior Técnico para a Investigação e Desenvolvimento

Additional collaborators
Susana Pereira da Silva, Instituto Nacional de Saúde Doutor Ricardo Jorge (INSA)
Rui Fragoso, Portuguese Energy Agency (ADENE)

Abstract

Space is growing in economic and strategic importance as launch access becomes cheaper and more frequent. However, significant barriers to entry still exist for both universities and first-time satellite builders in industry: Commercially available hardware is too expensive, and a lack of hardware and software interoperability mean that a substantial amount of engineering effort is required to integrate components. This project will develop Astrobit: a fully integrated open-source avionics and software ecosystem for CubeSats that will lower barriers to entry, enable faster and cheaper small satellite missions, foster collaboration, and increase mission success rates. Astrobit will also be tightly integrated into educational initiatives, including a space systems summer school whose curriculum will be released freely online.

MIT PI
Zac Manchester, Professor, Department of Aeronautics and Astronautics

PT PI
João Paulo Monteiro, Researcher, Associação do Instituto Superior Técnico para a Investigação e o Desenvolvimento - Instituto de Sistemas e Robótica 

Additional collaborator
Paul Fucile, Lecturer, Department of Aeronautics and Astronautics, MIT

Abstract

This project investigates the integration of computational 3D knitting and EBAM to develop novel lightweight architectural metamaterial systems for sustainable construction. Producing materially efficient horizontal building system (i.e., floorslabs) with current large-scale EBAM approaches remains constrained by geometric limitations and integration of structural reinforcement. The research combines MIT’s expertise in computational 3D knitting, streamlined fabrication pipelines, and material-informed design methodologies with the University of Minho’s expertise in robotic fabrication, 3D concrete printing, hybrid manufacturing, and computational design-to-fabrication workflows. Herein, we propose the development of a hybrid textile–mineral fabrication systems where knitted reinforcement and textile membranes act as active components within EBAM workflows enabling adaptive, materially-differentiated, low-carbon construction systems [1,2]. Through iterative multi-scale prototyping and lifecycle-informed evaluation, we will develop new hybrid fabrication methodologies, computational workflows, and proof-of-concept (PoC) demonstrators to advance circular , lightweight, and resource-efficient architectural scale production.

MIT PI
Mariana Popescu, Professor, Department of Architecture

PT PI
Bruno Figueiredo, Associate Professor , Landscapes Heritage and Territory Laboratory, School of Architecture, Art and Design, University of Minho

Additional collaborators
Mich Lin, Ph.D. Candidate, Department of Aeronautics & Astronautics, MIT
Paulo J.S. Cruz, Full Professor, Landscapes Heritage and Territory Laboratory, School of Architecture, Art and Design, University of Minho
Filipe Brandão, Assistant Researcher , Landscapes Heritage and Territory Laboratory, School of Architecture, Art and Design, University of Minho

Abstract

Modern aircraft sustainment, a major Portuguese industry with national aeronautics MRO turnover estimated above €500M annually, faces growing challenges in repairing high-value metallic aerostructures. Conventional repairs, including mechanically fastened or bonded patches, can increase structural weight, extend downtime, and raise lifecycle operating costs. Friction stir deposition (FSD) offers a promising alternative by rebuilding damaged material locally with low heat input and reduced distortion. However, unlocking FSD as an aviation repair technology requires predictive, geometryaware methods for co-optimizing process parameters and toolpaths. To this end, this project will develop a physics-informed AI framework for FSD repair of curved aluminum aerostructure elements. By combining 3D surface characterization, process physics, and generative toolpath optimization, the work will create repair strategies that improve thermal control, deposition quality, and fatigue performance. These results will lay the foundation for AI-driven, certifiable aerospace repair by linking repair planning, process physics, and structural performance in a traceable framework.

MIT PI
Zachary Cordero, Professor,  Department of Aeronautics and Astronautics

PT PI
Eduardo A. S. Marques, Associate Professor of Mechanical Engineering,
University of Porto, INEGI - Instituto de Ciência e Inovação em Engenharia Mecânica e Engenharia Industrial

Abstract

Building on a recently demonstrated photopatterned thiol-ene-epoxy resin capable of over three orders of magnitude variation in stiffness from a single precursor formulation, GradJoint will establish design rules for two-dimensional gradient adhesive architectures that maximize joint strength and stretchability between disparate substrates (glass, polycarbonate, acrylic, silicone, polyurethane). The collaboration uniquely couples MIT's expertise in photochemistry and multimaterial manufacturing with INEGI's three-decade track record in structural adhesive joints and a newly developed micro-scale simulator that connects adhesive flow, cure kinetics and interfacial mechanics to create a unified computational experimental workflow. The project will quantify adhesion across disparate substrates, develop predictive models linking photopatterned modulus fields to stress redistribution and joint performance, and experimentally validate optimized gradient architectures in a wearable electronics demonstrator. The resulting open design framework will enable digitally programmable adhesive interfaces tailored to arbitrary material pairings and device geometries.

MIT PI
T. j. Wallin, Professor,  Department of Materials Science and Engineering

PT PI
Lucas F. M. da Silva, Full Professor, Department of Mechanical Engineering, Faculty of Engineering of University of Porto

Abstract

Recent extreme events, such as Storm Kristin on 28 January 2026, have highlighted significant vulnerabilities in the National Electricity System (SEN), resulting in disruptions to transmission and distribution networks which affected security of supply. This project aims to provide the SEN with robust analytical tools to support public policies and structural investments, aligned with the objectives of climate resilience, modernization of critical infrastructure and the energy transition. The focus is on: a) Assessment of structural measures to strengthen the resilience of the SEN, considering current and potential climate scenarios; b) Analysis of the technical feasibility and economic rationale for the selective undergrounding of electricity lines in areas identified as critical; c) Formulation of proposals for adapting electricity network planning instruments to the new climate reality, including the integration of risk and resilience metrics into decision-making processes; and, d) Assessing relevance of the methods for resilient electricity services in the US.

MIT PI
Marija Ilic, Senior Research Scientist, Adjunct Professor,  Laboratory for Information and Decision Systems, Department of Electrical Engineering and Computer Science

PT PI
Pedro Carvalho, Full Professor, Institute for Systems and Computer Engineering, University of Lisbon (IST)