Machine Learning Software Engineer

Flagship Pioneering•Cambridge, MA

8d•$96,000 - $156,750

About The Position

Mirai Bio, Inc is a Flagship backed, privately held biotechnology company on a mission to transform the current approach to information molecule therapeutics to unlock their full therapeutic potential. In recent years, we have begun to experience the power of information molecules in treating historically undruggable diseases and in designing therapies with unprecedented turnaround times. Mirai’s platform integrates nanoparticle development with world-class informatics technologies and a novel pipeline of experimentation and discovery to drive a new generation of highly effective, therapeutically relevant information molecule therapies. We are seeking collaborative, relentless problem solvers that share our passion for impact to join us! Mirai was founded by Flagship Pioneering. Flagship Pioneering conceives, creates, resources, and develops first-in-category life sciences companies to transform human health and sustainability. Since its launch in 2000, the firm has applied a unique hypothesis-driven innovation process to originate and foster more than 100 scientific ventures, resulting in over $30 billion in aggregate value. The current Flagship ecosystem comprises 37 transformative companies, including: Moderna Therapeutics (NASDAQ: MRNA), Rubius Therapeutics (NASDAQ: RUBY), Indigo Agriculture, and Sana Biotechnology (NASDAQ: SANA). The Role Mirai Bio is seeking a highly talented individual to design, implement, and deploy novel ML approaches to optimize in vivo therapeutic delivery vehicles. They will work cross-functionally to build the computational tools necessary to identify and test therapeutic candidates to enable Mirai’s next generation genomic medicines. The successful candidate will engineer scalable ML platforms and pipelines to overcome limitations of current nucleic acid delivery approaches and will have a strong understanding of targeted information molecule delivery. A successful candidate will have strong familiarity with production-grade ML systems, active learning, and experience working with chemoinformatic libraries. The candidate will also be expected to collaborate extensively with experimental and infrastructure teams.

Requirements

PhD or Master's in Computer Science, Applied Mathematics, Bioengineering, Chemical Engineering, or a related quantitative field with a strong ML and Software Engineering focus.
4+ years of experience developing and deploying ML models in a production environment within industry and/or academic settings.
Strong experience with uncertainty quantification, active learning, and Bayesian Optimization in drug delivery, materials science, or related fields.
Proficiency in ML frameworks (PyTorch/TensorFlow/JAX) and the Python data science ecosystem.
Demonstrated experience with MLOps principles (version control, CI/CD, monitoring) and hands-on experience with cloud computing infrastructure (e.g., AWS, GCP, Azure) to accelerate model training, deployment, and inference.
Strong independent problem-solving ability and attention to detail.
Demonstrated achievement in industry or academia (publications, patents, or successful ML system deployments).
Excellent communication and presentation skills for both technical and interdisciplinary audiences.
Enthusiasm for working with cross-functional teams of experimentalists, engineers, and computational scientists in a fast-paced, entrepreneurial environment.

Nice To Haves

Hands-on experience building and managing ML packages.
Experience with ML experiment tracking platforms (e.g. MLflow, Weights & Biases)
Familiarity with API development and microservices for integrating ML models into experimental workflows.
Background in drug delivery, lipid chemistry, or nanoparticle formulation.

Responsibilities

Design, build, and deploy production-grade ML models for optimization of lipid nanoparticle (LNP) formulation, synthesis, and in vivo performance.
Develop and implement scalable software best practices for uncertainty quantification, ensuring models are reliable for real-time decision-making.
Engineer robust data pipelines to integrate multi-fidelity datasets (in silico and in vivo) to accelerate data-driven discovery of novel LNPs
Collaborate with computational scientists to translate research models into maintainable, efficient software that identifies design pathways for LNPs that achieve targeted functional properties
Work with infrastructure and automation teams to architect and streamline real-time data transfer between predictive models and experimental platforms
Partner with experimental teams to drive iterative design–make–test–analyze (DMTA) cycles
Communicate findings to stakeholders and leadership through written reports and technical presentations.