What are the responsibilities and job description for the Applied ML Scientist position at Powerhouse Biology?
About Powerhouse Biology, Inc.
Powerhouse Biology, Inc. is a venture-backed biotechnology company. We are pioneering precision protein and peptide therapeutics to address mitochondrial dysfunction, an underlying driver of many age-related diseases. Leveraging AI-guided multi-modal analysis, we develop cutting-edge therapeutics that restore mitochondrial function with unprecedented precision.
Join a fast-moving startup tackling one of biology’s biggest challenges: advancing healthy human lifespan. Work at the intersection of systems biology, synthetic biology, and machine learning, collaborate with creative and bright minds, and build systems that directly impact therapeutic development. If you thrive in a high-impact, no-nonsense environment where we innovate together to drive results, we want you on our team!
The role
We are seeking a creative and rigorous Applied ML Scientist to build computational frameworks for multi-modal biological data integration. You’ll lead the development of the learning loop that connects complex, multi-modal biological data to actionable hypotheses. You’ll compose and evaluate vision, sequence, and graph models with clear statistical judgment, encode mechanistic priors grounded in mitochondrial biology, and iterate quickly with prospective feedback from the lab. We’re looking for someone comfortable in ambiguity, opinionated yet data-driven, who builds rigorously and fast, communicates crisply across disciplines, and thrives in a high-agency startup. You’ll work across data engineering, systems biology, and machine learning to shape how we model complex disease mechanisms and surface tractable interventions. The ideal candidate is an inventive, low-ego collaborator who celebrates interpretability, calibration, and reproducibility as much as performance.
Key responsibilities
- Design and maintain multi-modal ML workflows for ingestion and processing of imaging and omics data.
- Write reliable, readable code; document assumptions; version data and models; contribute to simple, durable team standards.
- Help shape the technical culture through thoughtful code review, knowledge sharing, and pragmatic tooling choices.
- Partner with the experimental team to frame questions, scope datasets, and plan iterative validation to translate computational insights into real-world therapeutics.
Who you are
- Quantitative background (CS, applied math, physics, engineering, computational biology, or similar) with 2-5 years of hands-on ML for scientific or complex data.
- Strong Python and modern ML frameworks; comfortable with at least two of: computer vision, tabular/omics modeling, graph methods.
- Solid statistical instincts (splits, controls, basic calibration) and an eye for clean, reproducible work.
- Demonstrated expertise working with high-content cell culture image data and at minimum one type of omics dataset in an independent or lead capacity, such as driving analysis pipelines, integrating multi-omics data, or publishing results.
- Skilled with network modeling, graph algorithms, and computational approaches to systems biology, including building, analyzing, and interpreting biological graphs.
- Comfortable working in a collaborative, interdisciplinary environment, integrating computational methods with experimental biology.
- Able to balance research and engineering, delivering robust computational tools while exploring novel methodologies.
- Effective at communicating in multi-disciplinary teams where you are simultaneously the expert and the apprentice.
Bonus qualifications
- Familiarity with cloud computing and high-performance computing environments.
- Experienced in software development best practices, including version control (e.g. Git), code review, testing frameworks, workflow orchestration, and CI/CD pipelines.
- Prior work translating model interpretability into experimental design changes.
- Knowledge of mitochondrial biology.