Scientist I, AI Engineer

Frontier Medicines·
South San Francisco, California, United States
Yesterday
Full-timeITSmall Molecule
$135K - $159K/yr(estimated)

Description

<p>Frontier Medicines is&nbsp;seeking&nbsp;a highly motivated&nbsp;Scientist,&nbsp;AI&nbsp;Engineer&nbsp;to join our&nbsp;AI&nbsp;organization. This role will play a key scientific and technical role in advancing Frontier’s&nbsp;AI&nbsp;and cheminformatics capabilities in support of&nbsp;covalent&nbsp;small molecule drug discovery.&nbsp;</p><p>The successful candidate will design and deploy machine learning models, agentic AI systems, and multimodal foundation&nbsp;and generative&nbsp;models that&nbsp;leverage&nbsp;Frontier's large-scale covalent chemistry and chemoproteomics data to accelerate compound design, prioritization, and optimization. This role works closely with medicinal chemistry, biology, and cross-functional drug discovery teams to build AI-powered tools that translate complex, heterogeneous data into actionable scientific insight&nbsp;</p><p>This position is ideal for an individual contributor who thrives in scientifically uncharted territory and enjoys building novel solutions&nbsp;for&nbsp;medicinal&nbsp;chemistry, biology, and computational chemistry teams.&nbsp;</p><p>This is an exciting opportunity in our South San Francisco site to deploy&nbsp;AI&nbsp;to make a difference for patients suffering from debilitating diseases by working in a highly collaborative and energetic team in a startup environment with short communication lines across functions and departments&nbsp;&nbsp;</p><p>&nbsp;</p> <p><strong>What we are looking for: </strong></p><p></p><ul><li>Design architecture and training strategies for domain-specific multimodal&nbsp;AI&nbsp;models that reason&nbsp;over&nbsp;molecular structures, assay data, protein structures, and scientific text — trained on Frontier’s proprietary data.&nbsp;Familar&nbsp;with&nbsp;self-supervised learning, and&nbsp;Mixture of experts (MoE)&nbsp;and&nbsp;&nbsp;Distributed&nbsp;training (DeepSpeed, FSDP)&nbsp;</li></ul><ul><li>Design and build multi-agent systems for scientific research workflows — supervisor/sub-agent orchestration,&nbsp;MCPs,&nbsp;task decomposition, tool calling, error recovery, and human-in-the-loop checkpoints&nbsp;that interface with internal systems&nbsp;(scientific&nbsp;databases,&nbsp;&nbsp;compound&nbsp;registries)&nbsp;</li></ul><ul><li>Develop and adapt generative molecular design approaches — diffusion models, autoregressive transformers, VAEs, reinforcement learning–guided generation — to explore and expand covalent compound libraries and novel chemical space.&nbsp;&nbsp;</li></ul><ul><li>Build multimodal RAG pipelines over structured data, slide decks, experimental reports, and images&nbsp;&nbsp;</li></ul><ul><li>Train, evaluate, and deploy deep learning models for molecular property prediction, virtual screening, ADMET modeling, and SAR analysis on Frontier’s covalent chemistry data.&nbsp;</li></ul><ul><li>Implement and benchmark molecular representations — GNNs, equivariant architectures, learned fingerprints — and build multi-task,transfer&nbsp;learning, and active learning&nbsp;approaches that&nbsp;leverage&nbsp;proprietary data&nbsp;for low-data regime predictions on active programs.&nbsp;</li></ul><ul><li>Deploy models into serving infrastructure that researchers and agents can query — endpoints, batch inference, monitoring — so predictions&nbsp;reach&nbsp;decisions.&nbsp;&nbsp;</li></ul><p>&nbsp;</p><p><strong>Traits we believe make a strong candidate:&nbsp;</strong></p><p>&nbsp;</p><ul><li>B.Sc&nbsp;in computational or quantitative discipline such as&nbsp;computer science, data science,&nbsp;cheminformatics, computational&nbsp;chemistry&nbsp;or&nbsp;related field with 2-3 years industry experience. </li></ul><ul><li>Strong programming skills in&nbsp;Python&nbsp;and experience with coding tools.&nbsp;</li></ul><ul><li>Ability to work independently while thriving in a highly collaborative, cross-functional environment.&nbsp;</li></ul><ul><li>Excellent written and verbal communication skills.&nbsp;</li></ul><ul><li>Legally authorized to work in the United States.&nbsp;</li></ul><p></p><p><strong>Leveling Guidelines</strong></p><p><strong>Scientist I</strong></p><ul><li> <strong>Education &amp; Experience:</strong> </li><ul><li> Bachelor’s degree with 2–3 years of relevant industry experience, or </li><li> PhD with at least 1 year of relevant industry experience </li></ul><li> Demonstrated ability to independently or with limited guidance design and execute computational and AI approaches for drug discovery problems </li><li> Proven track record of contributing to complex analytical strategies and influencing discovery programs through data-driven insights </li><li> Ability to effectively collaborate with experts across disciplines and contribute to cross-functional team success</li></ul><p>&nbsp;</p> <ul><li>Health Care Plan (Medical, Dental &amp; Vision)</li><li>Retirement Plan (401k, IRA)</li><li>Life Insurance (Basic, Voluntary &amp; AD&amp;D)</li><li>Paid Time Off (Vacation, Sick &amp; Public Holidays)</li><li>Family Leave (Maternity, Paternity)</li><li>Short Term &amp; Long Term Disability</li><li>Training &amp; Development</li><li>Free Food &amp; Snacks</li><li>Wellness Resources</li><li>Stock Option Plan</li></ul><p></p><p><em>At Frontier, we strive to build a diverse and equitable workplace. The salary range for this role is $134,840 - $158,900. Compensation for the role will depend on a number of factors, including candidates' qualifications, skills, competencies and experience. Frontier offers a competitive total rewards package which includes healthcare coverage, 401k and a broad range of other benefits. </em></p><p><em>This compensation and benefits information is based on Frontier's knowledge as of the date of publication, and may be modified in the future. </em></p>

Requirements

<p><strong>What we are looking for: </strong></p><p></p><ul><li>Design architecture and training strategies for domain-specific multimodal&nbsp;AI&nbsp;models that reason&nbsp;over&nbsp;molecular structures, assay data, protein structures, and scientific text — trained on Frontier’s proprietary data.&nbsp;Familar&nbsp;with&nbsp;self-supervised learning, and&nbsp;Mixture of experts (MoE)&nbsp;and&nbsp;&nbsp;Distributed&nbsp;training (DeepSpeed, FSDP)&nbsp;</li></ul><ul><li>Design and build multi-agent systems for scientific research workflows — supervisor/sub-agent orchestration,&nbsp;MCPs,&nbsp;task decomposition, tool calling, error recovery, and human-in-the-loop checkpoints&nbsp;that interface with internal systems&nbsp;(scientific&nbsp;databases,&nbsp;&nbsp;compound&nbsp;registries)&nbsp;</li></ul><ul><li>Develop and adapt generative molecular design approaches — diffusion models, autoregressive transformers, VAEs, reinforcement learning–guided generation — to explore and expand covalent compound libraries and novel chemical space.&nbsp;&nbsp;</li></ul><ul><li>Build multimodal RAG pipelines over structured data, slide decks, experimental reports, and images&nbsp;&nbsp;</li></ul><ul><li>Train, evaluate, and deploy deep learning models for molecular property prediction, virtual screening, ADMET modeling, and SAR analysis on Frontier’s covalent chemistry data.&nbsp;</li></ul><ul><li>Implement and benchmark molecular representations — GNNs, equivariant architectures, learned fingerprints — and build multi-task,transfer&nbsp;learning, and active learning&nbsp;approaches that&nbsp;leverage&nbsp;proprietary data&nbsp;for low-data regime predictions on active programs.&nbsp;</li></ul><ul><li>Deploy models into serving infrastructure that researchers and agents can query — endpoints, batch inference, monitoring — so predictions&nbsp;reach&nbsp;decisions.&nbsp;&nbsp;</li></ul><p>&nbsp;</p><p><strong>Traits we believe make a strong candidate:&nbsp;</strong></p><p>&nbsp;</p><ul><li>B.Sc&nbsp;in computational or quantitative discipline such as&nbsp;computer science, data science,&nbsp;cheminformatics, computational&nbsp;chemistry&nbsp;or&nbsp;related field with 2-3 years industry experience. </li></ul><ul><li>Strong programming skills in&nbsp;Python&nbsp;and experience with coding tools.&nbsp;</li></ul><ul><li>Ability to work independently while thriving in a highly collaborative, cross-functional environment.&nbsp;</li></ul><ul><li>Excellent written and verbal communication skills.&nbsp;</li></ul><ul><li>Legally authorized to work in the United States.&nbsp;</li></ul><p></p><p><strong>Leveling Guidelines</strong></p><p><strong>Scientist I</strong></p><ul><li> <strong>Education &amp; Experience:</strong> </li><ul><li> Bachelor’s degree with 2–3 years of relevant industry experience, or </li><li> PhD with at least 1 year of relevant industry experience </li></ul><li> Demonstrated ability to independently or with limited guidance design and execute computational and AI approaches for drug discovery problems </li><li> Proven track record of contributing to complex analytical strategies and influencing discovery programs through data-driven insights </li><li> Ability to effectively collaborate with experts across disciplines and contribute to cross-functional team success</li></ul><p>&nbsp;</p>
Frontier Medicines

Frontier Medicines

BIOTECHNOLOGY

Drug Discovery Using Chemoproteomics Platform

LocationCA - South SF
Open Jobs3
Oncology
View Company Profile

Pipeline

FMC-376Phase 1/2