Senior / Staff ML Scientist
Machine learning lies at the core of insitro’s approach to rethinking drug development. Our team is responsible for the small molecule machine learning platform at insitro, and our responsibilities include design of high-throughput data generating experiments, design and execution of ML driven solutions for accelerating key problems in the drug discovery space, and creation of APIs and tooling for making the entire process reproducible and faster over time.
We are rapidly developing our small molecule hit-finding capabilities through cutting-edge machine learning technology. One of the key technologies we develop and leverage at insitro is DNA encoded libraries (DELs). Using DELs, we rapidly generate drug hits for validated targets coming out of our platform. Our combination of automated screening technology and machine learning enables us to extract promising hits from typically noisy DEL datasets and iterate far more rapidly than traditional high throughput screening approaches.
In this role, you will develop new models for molecule optimization utilizing data from multiple modalities, including DNA encoded libraries. To be effective, these models require powerful molecule representations that exploit relevant chemical principles and synthesize data from disparate sources. To achieve this, you’ll partner directly with chemists, machine learning scientists, hardware and software engineers, and clinicians to drive our drug discovery programs.
What you’ll do day to day:
- Deep dive into our DNA encoded libraries to define molecular models that incorporates the correct inductive biases to leverage the huge combinatorial datasets of DELs
- Collaborate cross-functionally with folks from our machine learning, automation, chemistry, and especially computational chemistry groups to develop new machine learning tools for hit discovery and lead optimization
- Directly shape the roadmap for our ChemML platform
- Shape our team’s culture with your ideas and experience
- All the normal ML stuff: document your method development, implement and share your methods in code, write and review design docs, talk to collaborators, and do code reviews
- Ultimately you’ll move the needle in a meaningful way for insitro and the field of medicine!
Examples of projects you will be working on:
- Design new model architectures that leverage knowledge about the chemical building blocks used to synthesize our DNA encoded libraries
- Leverage novel generative methods to generate new molecules in 2D and 3D space
- Better understand and develop machine learning methods for making out-of-distribution predictions for molecular datasets
- Collaborate with our software engineering teams to streamline dataset generation for model training
- Collaborate with our clinical scientists to deploy your models to accelerate a specific therapeutic area pipeline
In return, we will support you by:
- Placing a high degree of trust in your ideas and execution
- Bringing you up to speed on our DNA encoded library technology
- Making ourselves available for collaboration
- Caring about you as a whole person - not a resource
- Being a well funded startup with conservative runway
- You have 5+ years of experience as a machine learning scientist, with 2+ years of experience with chemistry datasets
- You’re eager to ship work that makes a difference to scientists and ultimately patients
- You feel comfortable reasoning about the tradeoffs between quality and speed when building in a startup environment
- You have experience with at least one high-end ML development environment such as Tensorflow or Pytorch
- You have experience with at least one cheminformatics toolkit such as OpenEye, RDKit, or Schrodinger
- You’ve demonstrated your ability to develop novel machine learning methods that go beyond putting together existing code, and to apply problem-solving skills to complex issues in the chemistry domain
- Experience with protein-ligand binding datasets
- Experience with DNA encoded library datasets
- Experience with our software stack: AWS, python, SQLAlchemy, PostgreSQL, Docker, workflow engines such as redun, pytorch
Compensation & Benefits at insitro:
Our target starting salary for successful US-based applicants for this role is $170,000 - $225,000. To determine starting pay, we consider multiple job-related factors including a candidate’s skills, education and experience, market demand, business needs, and internal parity. We may also adjust this range in the future based on market data.
This role is eligible for participation in our Annual Performance Bonus Plan (based on company targets by role level and annual company performance) and our Equity Incentive Plan, subject to the terms of those plans and associated policies.
In addition, insitro also provides our employees:
- 401(k) plan with employer matching for contributions
- Excellent medical, dental, and vision coverage (insitro pays 100% of premiums for employees), as well as mental health and well-being support
- Open, flexible vacation policy
- Paid parental leave
- Quarterly budget for books and online courses for self-development
- Support to occasionally attend professional conferences that are meaningful to your career growth and development
- New hire stipend for home office setup
- Monthly cell phone & internet stipend
- Access to free onsite baristas and cafe with daily lunch and breakfast
- Access to free onsite fitness center
- Commuter benefits