Curriculum vitae
Ian M. Pendleton, Ph.D.
Objective
To accelerate the design of new medicines for unmet medical needs across druggable chemical space through closed-loop experimentation under human supervision — coupling synthetic route design and medicinal chemistry judgment with machine learning, virtual screening, and laboratory automation so that each experiment trains the next design cycle.
Skills
Chemistry
Parallel and combinatorial library design; reaction condition development; high-throughput experimentation and robotic synthesis; retrosynthesis and route planning; CRO synthesis direction; SAR analysis (including Free–Wilson) and multi-parameter property optimization; LC-MS and HPLC data processing and interpretation; NMR characterization. Trained as a synthetic organic chemist (B.S. and Ph.D.).
Modeling & design
Structure-based drug design; molecular docking; free energy perturbation (FEP); docking tool and workflow buildout; ultra-large virtual screening with in-house 3D shape matching; generative design (REINVENT4, Chemistry42, Iktos Makya) tuned with MPO scoring (MOEsaic, CCG); Pareto-front exploration; Bayesian optimization.
Machine learning & AI
Expert-guided featurization; multi-task model development and assessment; conformal prediction and domain-of-applicability analysis; assay data curation; Python (RDKit, pandas, scikit-learn, PyTorch); API design for molecular modeling services (AWS, Prefect, Pulumi); Model Context Protocol (MCP) server development and agentic workflows.
Industry experience
Associate Principal Scientist, Cheminformatics & AI Design — Alkermes, Inc.
Senior Scientist — Alkermes, Inc.
- Contributed directly to the lead series on both programs supported since joining; ultra-large virtual screening broke out of existing patent space, establishing new series built on previously unreported analogs.
- Developed in-house 3D shape-matching search for ultra-large virtual screening (60+ billion molecules assessed), combined with docking and machine learning scoring to nominate potent, synthetically tractable compounds.
- Direct parallel synthesis campaigns with CRO partners: library design and building-block selection from established chemistry with CRO and internal medicinal chemists, identifying promising intermediates and proposing alternative scaffolds; advocate with partners for reaction-focused machine learning APIs; work up returned LC-MS and assay data into SAR.
- Drive multi-parameter optimization: tune generative models (REINVENT4, Chemistry42, Iktos Makya) against MPO scores, explore Pareto fronts across potency and physicochemical properties, and apply conformal prediction to support decisions in fast-moving, data-limited SAR.
- Leading development of agentic AI workflows: built internal Model Context Protocol (MCP) servers for modeling tools and use MCP-connected agents to accelerate established medicinal chemistry workflows (Free–Wilson analysis, library design).
- Built and operate full-stack machine learning and molecular modeling services, including workflow tools for model error evaluation and retraining; introduced active learning to Alkermes design cycles; partner with medicinal chemists on synthesis planning, hit-finding, and lead-optimization strategy.
Senior Machine Learning Scientist — Relay Therapeutics
Machine Learning Scientist II — Relay Therapeutics
- Ran direct-to-assay library synthesis end to end — virtual screening and library design, building-block selection, reaction condition development, robotic synthesis execution, and compound registration — reducing design–make–test cycle time from 50 days to ~7 days.
- Wrote processing code automating analysis of thousands of LC-MS and HPLC runs for DNA-encoded library (DEL) screening and direct-to-assay parallel synthesis, saving the DEL team months and helping it meet critical company milestones following the ZebiAI acquisition.
- Built the supporting software (Python, scikit-learn, PyTorch, Pulumi, Prefect): an API-driven application for library synthesis requests, automated dispense worklists, mass spectrometry inputs, and registration; molecule selection via vector-based enumeration and generative models, with active learning on assay data introduced for the first time at Relay.
- Initiated and led a multi-university collaboration on reactivity-network-guided parallel library synthesis — secured funding, designed the automation and reaction conditions, wrote the instrument control code, generated early-phase synthesis data, and drafted the initial manuscript.
Research Scientist I — Vertex Pharmaceuticals
- Built reaction platforms alongside medicinal chemists; collected and analyzed multistep synthetic route data to guide route selection and reaction conditions.
- Led machine-learning reaction prediction and retrosynthesis efforts; built a web application (Python, AWS) for reaction search, success statistics, and recommendation over internal and patent reaction data.
- Designed a multi-parameter Bayesian optimizer for high-throughput chemistry, reducing the experiments required for optimization by ~60%.
Education & training
Cheminformatics Postdoctoral Fellow — Haverford College
- Led and organized a materials discovery campaign across six research institutions and 15+ scientists in a DARPA consortium; developed its machine learning and cheminformatics pipelines and helped build chemistry capability at Emerald Cloud Lab for multi-site reproducibility.
Ph.D., Organic Chemistry — University of Michigan
- Synthesized and characterized cobalt complexes for CO2 reduction catalysis; related ligand structure to activity through paired QM and QSAR analysis.
- Computational investigation of C(sp3)–N, C(sp2)–C(sp3), and C(sp3)–F reductive elimination from Pd(IV) (JACS 2016).
- Published work applying automated potential energy surface and transition-state exploration methods (Q-Chem, Python, C++) to experimentally relevant Pd systems and CO2 reduction chemistry.
B.S., Professional Biochemistry — Eastern Michigan University
- Synthetic thesis: “Optimization of the aza-Cope rearrangement–Mannich cyclization reaction of conformationally mobile acylpyrrolidines.”
Publications
Also on Google Scholar and ORCID. ‡ equal contribution.
- Shim, E.; Dunstan, D. R.‡; Kong, Z.‡; Pendleton, I. M.‡; Zimmerman, P. M. Navigating Parallel Library Synthesis with a Reactivity Network. Chem, under review (revised, 2026).
- Estrada Pabón, J. D.; Haddox, H. K.; Van Aken, G.; Pendleton, I. M.; Eramian, H.; Singer, J. M.; Schrier, J. The Role of Configurational Entropy in Miniprotein Stability. J. Phys. Chem. B 2021, 125, 3057–3065.
- Pendleton, I. M.; Caucci, M. K.; Dharna, A.; Tynes, M.; Najeeb, M. A.; Chan, E. M.; Norquist, A. J.; Schrier, J. Untangling How Machines ‘Learn’ Perovskite Crystallization Chemistry. J. Phys. Chem. C 2020, 124, 13982–13992.
- Nisbet, M. L.; Pendleton, I. M.; Nolis, G. M.; Griffith, K. J.; Cabana, J.; Norquist, A. J.; Poeppelmeier, K. R. Machine-Learning-Assisted Synthesis of Polar Racemates. J. Am. Chem. Soc. 2020, 142, 7555–7566.
- Li, Z.; Najeeb, M. A.; Alves, L.; Sherman, A.; Parrilla, P. C.; Pendleton, I. M.; Zeller, M.; Schrier, J.; Norquist, A. J.; Chan, E. Robot-Accelerated Perovskite Investigation and Discovery (RAPID). Chem. Mater. 2020, 32, 5650–5663.
- Pendleton, I. M.; Cattabriga, G.; Li, Z.; Najeeb, M. A.; Friedler, S. A.; Norquist, A. J.; Chan, E. M.; Schrier, J. Experiment Specification, Capture and Laboratory Automation Technology (ESCALATE): A Software Pipeline for Automated Chemical Experimentation and Data Management. MRS Commun. 2019, 9, 846–859.
- Pendleton, I. M.; Pérez-Temprano, M. H.; Sanford, M. S.; Zimmerman, P. M. Experimental and Computational Assessment of Reactivity and Mechanism in C(sp3)–N Bond-Forming Reductive Elimination from Palladium(IV). J. Am. Chem. Soc. 2016, 138, 6049–6060.
- Li, M. W.; Pendleton, I. M.; Nett, A. J.; Zimmerman, P. M. Mechanism for Forming B, C, N, O Rings from NH3BH3 and CO2 via Reaction Discovery Computations. J. Phys. Chem. A 2016, 120, 1135–1144.
- Kolopajlo, L. H.; Hollis, N. K.; Pendleton, I. M. Kinetics of the Reaction between Ni(tetren)2+ and Bipyridine. Journal of Student Research 2012, 2, 39–45.
Other writing
- Pendleton, I. “The Greater Chemistry Community.” ACS Graduate and Postdoctoral Chemist, 2016, p. 15.
- Pendleton, I. “So, how about a Ph.D.?” ACS Undergraduate Blog (four-part series), 2012.
Software & data
- ESCALATE_Capture and ESCALATE_report — experiment capture and reporting for automated chemistry
- DARPA SD2 perovskite dataset — curated high-throughput crystallization data
Talks and posters are listed on the presentations page.
Awards & honors
- Karle Symposium Dow Sponsored Travel Award in Inorganic Chemistry, 2016
- Vaughn Symposium Dow Sponsored Travel Award in Organic Chemistry, 2013
- National Science Foundation Graduate Research Fellowship, 2013
- Rackham Merit Fellowship, University of Michigan, 2012
- ACS Division of Organic Chemistry Travel Award for Outstanding Undergraduate Students, 2012
- ACS Division of Organic Chemistry Summer Undergraduate Research Fellowship (Pfizer), 2011
- EMU Honors College Undergraduate Research Fellowship, 2010–2012
- EMU: Peet-Mayor Endowed Chemistry Award (2012), Biochemistry Achievement Award (2012), Honors Senior Thesis Award (2012), “Pete” Decoster Endowed Scholarship (2011), CAS Dean’s Travel Award (2011), Warren Scholarship (2011), Collins Endowed Scholarship (2010), Elwood J. Kureth Scholarship (2010), Sandra J. Lobbestael Endowed Scholarship (2009); Undergraduate Research Symposium Fellow, 2008–2012
Service & outreach
- Reviewer: J. Chem. Inf. Model., J. Chem. Theory Comput., Chemical Science, KDD Applied Data Science, and the ACS In Focus book Alternative Careers in Science
- Future Faculty Graduate Student Instructor, statistical mechanics and thermodynamics, 2017
- ACS Science Coaches Program, 2012–present
- Undergraduate mentoring through UROP, 2013–2015
- Washtenaw Elementary Science Olympiad coach, 2016
- President, EMU Chemistry Club (ACS Student Chapter), 2010–2012