Computational Scientist
Job description
Computational Scientist at Tamarind Bio
About the role
You will architect and deploy AI services that power structure prediction, protein design, and molecular docking inside production scientific workflows. Your primary mission is to convert exploratory research code into robust, scalable services that academic and industry researchers can depend upon. You will act as a bridge between platform technology and customer discovery challenges, ensuring that biological AI pipelines run smoothly and deliver measurable value.
Key facts
What you will do
Translate high-level discovery objectives into concrete model selection and pipeline architecture decisions.
Build end-to-end workflows that integrate structure prediction engines, docking simulations, and scoring functions.
Define and track performance and throughput metrics for scientific outputs, driving iterative improvements.
Lead deployment of services on cloud infrastructure, focusing on reliability and efficient scaling.
Engage with academic and pharmaceutical partners to pilot new capabilities and gather feedback.
Assess emerging research tools and determine their suitability for integration into the platform.
Maintain compute environments for molecular modeling, leveraging CUDA acceleration and isolated runtime contexts.
Support scientists in troubleshooting complex workflows and adopting docking and design frameworks correctly.
Analyze input data such as protein sequences and structural records to guide model configuration.
Document pipeline architectures and operational procedures so that workflows remain transparent and repeatable.
Requirements
Strong foundation in computational biology, computational chemistry, or bioinformatics, demonstrated by relevant education or research experience.
Familiarity with machine learning and physics-based methods applied to structural biology, virtual screening, and molecular simulations.
Hands-on experience handling biological data objects including three-dimensional structures, compound collections, and sequence databases.
Proficiency in Python and scientific computing paradigms, with a track record of writing maintainable, production-grade code.
Experience operating cloud platforms and modern machine learning stacks, with an emphasis on orchestrated services.
Residence in the San Francisco Bay Area or the ability to relocate there in a timely manner.
Nice to have
No additional requirements are specified at this time.
Skills and tools
Core languages and frameworks such as Python, PyTorch, TensorFlow, and CUDA.
Platform and packaging tools including Conda, Docker, and orchestration on AWS.
Cloud infrastructure components such as EC2, S3, and DynamoDB.
Specialized environments for molecular modeling, protein design, and structural biology applications.
Interface definitions and workflow orchestration capabilities that enable integration across tools.
Practical notes
Please verify all information on the official application page before proceeding.
Confirm that your experience aligns with both the explicit requirements and the day-to-day realities of the role.
Ensure that your submissions are complete and reflect your true capabilities.
About the role
You will own the design and delivery of AI tools for structure prediction, protein design, and docking inside production workflows. Your daily work turns scattered research code into reliable services for scientists. You communicate with clients to align Tamarind tools with their discovery goals.
Key facts
What you'll do
Orchestrating intake of client requirements and mapping them to model selection.
Designing build steps for protein structure prediction and molecular docking pipelines.
Leading review of scientific outputs against accuracy and throughput targets.
Coordinating ship of production workflows with reliable performance on AWS.
Establishing partnerships with academic and pharma collaborators to test platform features.
Evaluating new AI research tools for integration into the Tamarind platform stack.
Maintaining infrastructure for molecular modeling tasks using CUDA and Conda environments.
Troubleshooting customer pipelines and guiding correct usage of docking frameworks.
Analyzing biological data such as sequences and structures to inform model behavior.
Documenting workflow logic so operations remain clear and reproducible over time.
Requirements
You hold strong experience in computational biology, computational chemistry, or bioinformatics.
You are familiar with ML and physics-based tools for structural biology and virtual screening.
You have worked with biological data including molecular structures, compounds, and databases.
You write Python and manage scientific computing workflows with comfort.
You operate cloud infrastructure and ML tooling involving AWS, Docker, and PyTorch.
You are based in the SF Bay Area or can relocate there promptly.
Nice to have
None specified.
Skills and tools
Python, PyTorch, TensorFlow, CUDA, Conda, Docker, AWS, molecular modeling tools, protein design frameworks, structural biology tooling, APIs.