Machine Learning Scientist, Molecular Interactions at Terray Therapeutics, Inc.
$147,000–$227,850from the description
Jun 29, 2026
—
Jul 21, 2026
What this job asks for AI summary
This role centers on advancing machine learning models for molecular interactions in drug discovery — specifically pushing structure-informed generative models toward causal representations of in-vivo behavior. The work involves designing novel architectures to improve selectivity, generality, and accuracy of ML-guided intermolecular interactions, working with large proprietary datasets that feed directly into active drug programs. It suits a researcher with deep expertise in molecular ML modeling who is comfortable pursuing speculative, genuinely novel approaches.
Senior level · Remote · Full-time
Posted 2 times — it's one opening, so apply once.
We read this from the posting text with AI. Skim the description below before ruling yourself out.
How this req sits in the market our data
Roughly 23,000 people nationally plausibly meet what this posting asks for (data scientists). range 9,600–34,400
Applicant volume Heavy — This req sits in a large pool with little in its requirements to thin it, and auto-apply tools fire at everything in the occupation. Applying early and leading with the rare skills below is what gets read.
What the occupation pays Median $122,874 (middle half $87,544–$162,374). This posting is about at that midpoint.
Estimated from BLS employment for this occupation and area, per-skill prevalence across our listing corpus, and published wage benchmarks — as of Jul 28, 2026. It is a model, not a headcount.
Why we read it this way (9)
The role sits at the intersection of ML research and computational chemistry/drug discovery; 15-2051 (Data Scientists) was chosen over 15-1252 (Software Developers) because the primary work is novel ML modeling and scientific research rather than software engineering, but the runner-up is noted given the emphasis on model architecture development and training at scale.
No formal degree is stated as a hard requirement — the JD gates on demonstrated work (GitHub, proof of relevant work, or a one-page writeup) rather than a credential.
No overall years-of-experience figure is stated; seniority is assessed as Senior based on the requirement for 'mastery' of modern ML model development, comprehensive understanding of the field, and the expectation of independently pursuing novel research directions.
The title 'ML Scientist' carries no explicit level modifier, so the title states no level.
'Basic familiarity with ML Ops' and 'basic familiarity with other classes of ML models implicated in Drug Design' are listed under Required Qualifications, so they are treated as hard gates despite the softening 'basic familiarity' language.
GitHub is listed as part of the application gate ('only applicants with github, proof of relevant work, or a one-page writeup… will be considered'), making it a hard requirement for consideration; the alternatives (proof of relevant work, one-page writeup) are application-process substitutes rather than skill substitutes.
An 'opinionated perspective on the tension between the bitter lesson and physical constraints' is called out as 'a plus' — this is a domain-philosophy preference, not a named technology, so it is omitted from the skills list per extraction rules.
Ignored 4 non-technology phrase(s) as skills (responsibilities/concepts, not named tools): ML model development, molecular modeling, structure-based drug design, drug design ML models.
Caller marked this a fully-remote role — scored against the national candidate pool.
Read the full posting
The employer publishes the full description on their own site — read it there ↗. Or sign in to read it here — it's free, and it also lets you track this application.