Travis Kessler

Travis Kessler

PhD Computer Engineering · Full-stack · ML/AI · Scientific computing

Production software and ML systems grounded in published research — from scientific modeling and open-source tools to SaaS platforms and large-scale agentic analytics.

View workDownload CVGet in touch

About

AI and machine learning engineer with a record of impactful research, open-source software, and scalable AI systems for science, industry, and energy. I build production platforms and research tooling end to end across full-stack product work, ML/AI systems, and scientific computing.

  • Full-stack

    Cognitive Chemistry Labs (TypeScript, FastAPI, Supabase, Modal, Stripe, CI/CD); LLM/RAG demos.

  • ML/AI

    SCEPTER/DARPA agents and MLOps; transformers, RAG, and fine-tuning.

  • Scientific computing

    PhD and 12 publications; ECNet / ECabc; combustion, QSAR, and fuel property ML.

Education

  • PhD, Computer Engineering

    University of Massachusetts Lowell · May 2023

  • B.S., Computer Engineering

    University of Massachusetts Lowell · May 2018

Awards

  • Dean's Gold Medal for Outstanding Academic Achievement

    2023 · University of Massachusetts Lowell

  • Innovative Technology Solution, UMass Lowell DifferenceMaker 50K Idea Challenge

    2017 · University of Massachusetts Lowell

  • 1st place, Francis College of Engineering Prototyping Competition

    2016 · University of Massachusetts Lowell

More honors

  • Computer Engineering Department Award for Outstanding Ph.D.

    2023 · University of Massachusetts Lowell

  • Best Presentation, ASME ICEF 2019 Conference

    2019 · ASME

  • 1st place, Symbotic Warehouse Robot Prototyping Competition

    2018 · Symbotic

Experience

  1. Founding Technical Consultant

    – Present

    Cognitive Chemistry Labs

    • Engineered and deployed a highly scalable, serverless AI platform for computational chemistry and biology, accelerating high-throughput analysis and research workflows.
    • Built and launched a full-stack web application (TypeScript frontend, FastAPI backend), integrating Supabase for authentication and database management, and Modal for dynamic, serverless compute resources.
    • Delivered a user-centric website for account management, subscription billing, and compute job payments, integrating Stripe for secure, automated transactions.
    • Instituted comprehensive unit testing and CI/CD pipelines, driving code quality and enabling seamless automated deployment to development and production environments.

  2. Research Engineer

    –

    AIMdyn, Inc.

    • Spearheaded development of data-driven solutions leveraging generative AI and deep learning to optimize forecasting, simulation, and analytic engine performance for strategic planning and discovery.
    • Integrated scalable, reusable MLOps components (MLflow, CI/CD, containerization), expediting experimentation and deployment of machine learning models for simulation and COA validation.
    • Orchestrated server cluster workflows and job scheduling systems, enabling high-throughput simulation and analytic engine operations for large-scale scenario exploration.
    • Advanced the SCEPTER program by creating unscripted, goal-oriented agents that autonomously discover novel, relevant, and interpretable courses of action (COAs) in trusted military simulation environments.
    • Devised and implemented algorithms to manage exponential growth of the global state-action space, enabling rapid exploration of large-scale military scenarios and achieving scenario exploration speeds up to 100,000x real-time.

  3. Graduate Research Assistant

    –

    UMass Lowell Energy & Combustion Research Laboratory

    • Applied advanced machine learning and data engineering techniques for robust data preprocessing, modeling, and validation in alternative energy research projects.
    • Leveraged optimization algorithms (including biologically-inspired methods) to fine-tune hyper-parameters in supervised and unsupervised learning workflows, boosting predictive model accuracy.
    • Automated data preprocessing pipelines in Python, enhancing consistency, reproducibility, and accuracy for downstream statistical evaluation.
    • Mentored research teams and provided technical leadership on software development, ensuring quality through code reviews, collaborative troubleshooting, and version control best practices (Git).

  4. Implementation/DevOps Engineer

    –

    Valora Technologies

    • Engineered automated extraction workflows to convert complex legal, financial, and government documents into actionable data insights, improving client deliverables.
    • Designed and optimized robust ETL pipelines, streamlining client document processing from intake to final data delivery and reducing turnaround time.
    • Led onboarding and training for new Implementation/DevOps engineers, empowering teams to master advanced data mining, pipeline configuration, and DevOps best practices.

  5. Undergraduate Research Assistant

    –

    UMass Lowell Energy & Combustion Research Laboratory

    • Developed advanced predictive models to accurately forecast chemical properties from molecular structures, driving innovation in QSAR/QSPR research.
    • Designed and deployed high-performance neural network architectures for complex, multidimensional datasets, increasing model accuracy and reliability.
    • Created and released open-source software packages for machine learning, feature extraction, and hyper-parameter optimization, enabling the research community with robust, reusable tools.
    • Published impactful research in leading scientific journals and conferences, including Fuel, The Journal of Open Source Software, and the ASME Internal Combustion Engine Fall Conference.

Skills

Selected Work

  • Cognitive Chemistry Labs

    Cognitive Chemistry Labs makes advanced computational chemistry and drug discovery tools accessible to academia and small and mid-sized research companies. We provide affordable, transparent, and scalable AI tools to accelerate molecular research and the development of new solutions.

    • Full-stack
    • Scientific computing
    • TypeScript
    • FastAPI
    • Supabase
    • Modal
    • Stripe

  • ECNet

    QSPR-based machine learning for fuel property prediction — open-source tooling from the Energy and Combustion Research Laboratory.

    • ML/AI
    • Scientific computing
    • Python
    • QSPR
    • ML

  • PaDELPy

    A Python wrapper for PaDEL-Descriptor software, making molecular descriptor calculation accessible from Python workflows.

    • Scientific computing
    • Python
    • Cheminformatics

View all projects

Research

12 peer-reviewed publications spanning combustion, QSAR, and fuel property ML — selected works below.

  • Artificial neural network based predictions of cetane number for furanic biofuel additives

    Travis Kessler, Eric Sacia, Alexis Bell, J. Hunter Mack

    Fuel (2017) · DOI

    • ML
    • QSAR
    • Biofuels

  • A comparison of computational models for predicting yield sooting index

    Travis Kessler, Peter C. St. John, Junqing Zhu, Charles S. McEnally, Lisa D. Pfefferle, J. Hunter Mack

    Proceedings of the Combustion Institute (2020) · DOI

    • ML
    • Sooting
    • Combustion

  • ECNet: Large scale machine learning projects for fuel property prediction

    Travis Kessler, J. Hunter Mack

    Journal of Open Source Software (2017) · DOI

    • ML
    • Open source
    • Biofuels

  • Artificial Neural Network Models for Octane Number and Octane Sensitivity: A Quantitative Structure Property Relationship Approach to Fuel Design

    Amina SubLaban, Travis Kessler, Noah Van Dam, J. Hunter Mack

    Journal of Energy Resources Technology (2023) · DOI

    • ML
    • QSAR
    • Biofuels

  • CO2 and HDPE Upcycling: A Plasma Catalysis Alternative

    Fnu Gorky, Apolo Nambo, Travis J. Kessler, J. Hunter Mack, Maria L. Carreon

    Industrial & Engineering Chemistry Research (2023) · DOI

    • Plasma

View all research

Contact

Available for select work

Email meGitHubLinkedInBlueskyDownload CVBuy me a coffee