Senior Automation Engineer - High Throughput Experimentation

Satomic•San Diego, CA

1d•$120,000 - $140,000•Hybrid

About The Position

This role focuses on building automation and data tooling for synthetic chemists to develop and validate reactions, and owning process quality in the data generation pipeline. The responsibilities include translating experimental designs into executable automation (plate maps, worklists, instrument parameters), establishing controls and acceptance criteria for data quality, and troubleshooting issues within the physical process. The position is critical to Satomic's mission of accelerating the process from idea to molecule by ensuring experiments are well-designed and yield reliable data for ML models. The role requires a blend of scientific rigor and engineering skills, with a focus on collaboration with synthetic chemists and data scientists.

Requirements

Hands-on high-throughput experimentation experience, including scaling processes from low-throughput to high-throughput and setting process checks and acceptance criteria.
Fluency in design of experiments, including factorial and fractional designs, blocking, randomization, replication, power analysis, and variance splitting.
Proven ability to troubleshoot and identify root causes in physical processes.
Experience in teaching and scaling knowledge, leaving behind patterns and codifying intuition into SOPs or agents.
Ability to write scripts for data handling, reformatting instrument exports, cleaning spreadsheets, and integrating data between systems.
AI literacy and experience using AI tooling, with an interest in turning judgment into agent-executable processes.
Proficiency in Python and data skills to interrogate campaign data independently.
Sufficient scientific background to quickly understand synthetic chemistry and communicate with chemists as peers.

Nice To Haves

Experience in a fast-moving environment requiring planning and real-time reaction to incoming data.
Experience in a hybrid scientist and engineer role involving statistical rigor and bench/instrument time.
A proactive approach to identifying and fixing broken processes, even outside one's immediate scope.
Experience meeting domain experts (chemists, data scientists) where they are.
Experience influencing and raising the standard of work beyond one's direct ownership.
Experience building methods and being accountable for the data quality produced, not just the execution of the code.
Experience guiding colleagues who have not worked in a high-throughput setting.
Experience being opinionated, learning quickly, and driving change within a team.

Responsibilities

Own the automation tooling for synthetic chemists, building and refining it to bridge the gap between desired experiments and instrument capabilities.
Define the standards for data generation, including plate design, controls, replication, and validation criteria, ensuring experimental results are robust.
Implement and manage process quality control in the data generation pipeline, including controls, anchors, replication, and plate acceptance criteria to catch errors early.
Ensure execution quality for ongoing campaigns, monitoring adherence to standards, consistency across operators and time, and managing failures.
Identify and resolve process quality issues through side experiments, distinguishing real results from artifacts and pinpointing sources of bias.
Engineer new capabilities in the reaction process to enable previously impossible experiments and improve data reproducibility.
Map reaction development through the handoff into data generation, identifying and removing friction points.
Ship methods or tools for chemists' real work, learning from their actual usage.
Establish baseline metrics for data quality and reproducibility.
Create automated workflows supporting all steps of reaction development.
Identify and implement improvements to process quality and experimental results.
Manage reactions from development through to data generation campaigns with established standards.
Create a repeatable pipeline for onboarding and scaling reactions.
Develop parameterization and design tools that chemists use by default, encoding judgment into agent-executable processes.
Reduce campaign-level noise or material waste to improve information yield per well.
Make screening discipline native to the company's workflow, carried by a cohort of chemists and data-generation scientists.
Encode design judgment into agents that critique and correct campaign designs without human intervention.
Make challenging reactions scalable by solving development problems creatively and engineering reproducible solutions.