Senior Analytical Data Scientist

Satomic•San Diego, CA

3d•$165,000 - $180,000

About The Position

Satomic is hiring a Senior Analytical Data Scientist to build and own the models and workflows that convert high-throughput analytical chemistry instrument data into reliable labels describing the contents and concentrations of experimental wells. The role focuses on downstream computational analysis of raw data from Satomic’s small-molecule LC-MS workflows, including automated interpretation, data quality, uncertainty, and scalable processing of campaigns with diverse analytes and few authentic standards. The successful candidate will own the analysis pipeline from instrument output to label, including automated peak detection, product identification, adduct and fragment reasoning, feature annotation, conversion estimates, and routing ambiguous results for review. They will develop and maintain analytical QC policies, identify instrument faults and unusual chemistry, and ensure processing decisions and instrument and batch context are traceable. Longer-term work includes predicting analyte-specific detector response and retention time from chemical descriptors, and using those predictions to improve acquisition methods. This is a computational, domain-expert role rather than a bench, method-development, software-engineering, reactivity-prediction, or drug-design role. The scientist will set technical direction for how Satomic interprets instrument data, prototype as needed, and specify tools for software engineers to build. They will collaborate with the analytical chemistry team, cheminformatics and reactivity ML scientists, and software engineers. The position offers the opportunity to develop a new analytical data science function and advance high-throughput, standard-free quantitation.

Requirements

PhD plus 4 years, MS plus 7 years, or BS plus 10 years of experience in chemistry, analytical chemistry, data science, or a related field.
Strong fluency in analytical chemistry, particularly chromatography and mass spectrometry.
Experience processing raw chromatographic and mass spectrometry data, including writing or tuning code for peak picking, baseline correction, deconvolution, or feature detection.
Experience finding and fixing errors in an analytical data-processing pipeline.
Experience working directly with raw vendor files and understanding information lost when data is exported as a peak table.
Experience handling analog detector channels such as CAD and UV and moving their data into tools that can run in production.
Working knowledge of charged aerosol detection (CAD), including response-factor behavior and analytes it measures poorly.
Experience building automated analysis pipelines that process thousands of samples without manual triggering and scaling a pipeline through a large increase in volume.
Cheminformatics and data management exposure, including RDKit or equivalent for chemical structures, descriptors, and property prediction.
Experience designing and populating databases that link analytical results and processing parameters to raw data, samples, and batches.
SQL fluency, including writing queries that others use to retrieve data.
Rigor in calibration and uncertainty estimation, including providing error estimates for training labels.
Python fluency and the ability to prototype analytical tools.
Ability to specify software requirements precisely and work with engineers and coding agents to build tools, without being responsible for building production software.
Interest in predicting analyte behavior from chemical structure, including detector response, ionization efficiency, and retention time.

Nice To Haves

Experience building or running data pipelines in a modern untargeted or semi-targeted metabolomics or exposomics lab.
Molecular networking or MS/MS annotation of unknowns.
NMR and GC data analysis.
Recent hands-on bench experience.
Familiarity with synthetic organic chemistry.
Familiarity with Agilent and Waters LC-MS systems and their data formats, such as .d files.

Responsibilities

Build and own automated interpretation of raw instrument output, including peak detection, product identification, adduct and fragment reasoning, feature annotation, and conversion estimates with uncertainty.
Develop end-to-end pipelines that process full experimental campaigns unattended and scale analytical capacity without proportionally increasing manual review.
Own the analysis pipeline from raw instrument data to labels, deciding what runs automatically, what is flagged, and how failure modes are detected and addressed.
Determine which results are suitable for the reactivity ML team and which should be flagged or withheld; distinguish instrument faults from unusual chemistry.
Develop analytical quality-control policies and maintain traceability of processing choices, instrument context, and batch context.
Predict analyte-specific detector responses, including CAD and UV response factors, and retention times from chemical descriptors.
Use analytical predictions to inform acquisition strategies, including method length, gradients, internal standard placement, and multiplexed injections.
Guide the evolution of Satomic’s spectral visualization, annotation, and semi-automated review tools, working with software engineering resources.
Define the label contract with the reactivity ML team, including emitted data, uncertainty, and reasons for excluding wells.
Develop high-throughput feature annotation to identify likely side products and impurities and flag recurring unexpected products.
Establish measures of label quality and ground-truth experiments to validate models as chemistry expands.
Build the analytical data science function and, over time, the team that operates it.