Etinosa Osaro

Quantum Cheminformatics Scientist @ PsiQuantum

About

I build intelligent systems that execute science and engineering workflows end to end.I am a Chemical and Petroleum Engineer, AI/ML scientist, and full-stack software engineer working at the intersection of computational chemistry, quantum computing, and large-scale scientific infrastructure. My work focuses on designing systems that can autonomously explore, model, and optimize complex physical environments, from subsurface energy systems to molecular and materials discovery.My background spans reservoir engineering and production optimization, where I first applied machine learning to real-world oil and gas systems, through to advanced research in materials science and quantum chemistry. This foundation allows me to approach AI not just as a modeling tool, but as a mechanism for navigating high-dimensional physical design spaces grounded in real engineering constraints.Across academia and industry, I have developed integrated AI frameworks combining active learning, reinforcement learning, Gaussian processes, and deep learning to efficiently screen and discover materials, guide simulation campaigns, and reduce the cost of high-fidelity computation. These systems operate across Monte Carlo, molecular dynamics, and quantum chemistry environments, enabling data-efficient exploration of thermodynamic and molecular spaces at scale. At the frontier of this work, today, I design AI systems for chemistry on fault-tolerant quantum computers bridging classical and quantum systems for next-generation materials and molecular discovery. In parallel, I have built and led the development of production-grade software systems, including full-stack platforms, data pipelines, and machine learning infrastructure used in energy, analytics, financial, and engineering applications. I have managed engineering teams, deployed scalable applications, and translated complex models into usable products that operate in real-world environments.During my doctoral research, I developed foundational methods in active learning for adsorption and diffusion in porous materials, including sparse Gaussian process models, reinforcement learning-driven data acquisition, and universal adsorption modeling. I also built LLM-driven agentic frameworks that unify simulation, uncertainty quantification, and decision-making into cohesive, reproducible systems.

Country

City

United States

Industry

Research

Skill

Research and Development (R&D), Oil and Gas Drilling, Production Engineering, Deep Learning, Petroleum Engineering, Reservoir Simulation, Pump Design, Nonprofit Organizations, Leadership Management, Small Team Management, Building Leadership Teams, Web Applications, Web Design, Web Development, Mobile Applications, Oracle Database, Artificial Intelligence (AI), Active Learning, Monte Carlo Simulation, Monte Carlo modeling

Experience

Quantum Cheminformatics Scientist

PsiQuantum

2025-4 - Present · 1 yr 6 mos

Palo Alto, California, United States

AI/ML, Software Development, Molecular and Quantum Modeling for Fault Tolerant Quantum Computing.

Quantum Solutions Machine Learning

PsiQuantum

2024-6 - 2025-3 · 10 mos

Palo Alto, CA

AI in Quantum Chemistry, and Quantum Computing.

Doctoral Researcher

University of Notre Dame

2021-8 - 2025-3 · 3 yrs 8 mos

United States

* Developed active-learning algorithms to predict adsorption and diffusion isotherms across porous materials, significantly accelerating screening workflows * Implemented Gaussian Process Regression models to boost selectivity prediction accuracy for metal–organic frameworks (MOFs) * Created a sparse GPR package using inducing-point methods, enabling scalable regression on large materials datasets * Engineered custom reinforcement-learning tools (Q-learning & Proximal Policy Optimization) for automated data acquisition in active-learning campaigns * Designed and deployed a large-language-model (LLM)-based agentic framework to orchestrate molecular modelling and ML workflows * Developed a suite of Bayesian optimization acquisition-function libraries to streamline active-learning pipelines * Built a universal deep-learning model for adsorption prediction in MOFs, generalizable across MOF systems * Conducted Monte Carlo and molecular-modeling simulations to generate high-quality training data and validate predictive models * Developed machine-learning potentials for MOFs, improving fidelity of material simulations * Authored a Python package unifying molecular predictions with active-learning routines;

Manager, Software Development & Machine learning

Max-Migold

2020-8 - 2021-7 · 1 yr

* Led a software engineering team of five SWEs to deliver mission-critical applications on time and within budget * Architected and optimized the company’s facility-management platform and external website, improving system reliability and page-load performance * Designed and implemented an interactive data-visualization and analytics solution for a Stanbic IBTC engagement, enabling stakeholders to explore key metrics in real time * Developed a production-grade deep-learning model for Lagos State’s electrical-analysis project * Built and launched a cross-platform mobile and web application to streamline talent recruitment for partner businesses

Ambassador

Bank of America Merrill Lynch

2017-11 - 2019-8 · 1 yr 10 mos

Technical Intern (Production and Reservoir Engineering)

Oriental Energy Resources Limited

2018-8 - 2019-1 · 6 mos

Victoria Island Lagos, Nigeria

* Developed a machine-learning model to optimize production and predict performance for three oil wells experiencing Basic Sediment & Water (BS&W) challenges * Performed reservoir evaluation and built numerical models to characterize subsurface properties and inform development strategies * Compiled, cleaned, and analyzed production and reservoir data to identify performance trends and recommend operational improvements * Designed and implemented electric submersible pump (ESP) configurations, enhancing artificial-lift efficiency and well productivity