Computational Biology & Cheminformatics Expert

$80–110/hrrate as listed

micro1 is contracting computational biology and cheminformatics practitioners to build code-based benchmark tasks and review AI outputs for a customer's computational drug discovery program. Work involves curating bioactivity datasets, writing reproducible Dockerized tasks, and grading model reasoning on SAR, ADMET, and virtual screening problems.

Summary, eligibility guidance and screening preparation are prepared by Sidequest from the platform's posting. Check the source listing before applying.

What the work involves

You are writing the problems, not solving someone else's. A typical task starts with a realistic drug discovery scenario — filter a ChEMBL activity set for a target, flag assay inconsistencies, compute descriptors and cluster a series, triage a virtual screening hit list, or reason through an ADMET liability in a lead series — and ends with a runnable environment plus tests that verify a model actually got the science right. Expect to spend real time in Python and RDKit, in a terminal, and in Docker, then write the accompanying rationale explaining why a wrong answer is wrong.

The second half of the work is evaluation. You review AI-generated code, analyses, and chemistry rationales for scientific rigor: does the model understand that an IC50 from a different assay format isn't comparable, that a SMILES standardization step matters, that a QSAR model trained on that split is leaking? Feedback is written, specific, and cites the chemistry or biology — not "looks plausible."

What the screening looks for

Verifiable domain depth. Named targets, series, assays, and toolkits you have personally worked with, and what went wrong.
Actual coding ability. Python beyond a notebook script: packaging, CLI tools, pytest, Git, Docker. This is a genuine filter, not a preference.
Task-design judgment. Whether you can write a problem that is unambiguous, solvable, and has a defensible ground truth.
Calibration. Willingness to say "this is underdetermined" instead of grading confidently on a bad prompt.

Logistics

Fully remote, contractor, asynchronous. Contributors typically commit 10–20 hours per week with occasional scoped bursts; there are no fixed shifts, but tasks carry deadlines and rework requests. The $80–110/hr range reflects rates observed on this listing and varies with credentials, coding depth, and task type — treat it as observed, not guaranteed. Advanced degrees (PhD, MSc, PharmD) are strongly valued but industry track record can substitute.

What they look for

bioinformatics
biochemistry
ADMET
Cheminformatics
Molecular Docking
Structure-Activity Relationship
RDKit
Python
Git
GitHub
Docker
Test

Before you apply

Would you pass this screen?

Drawn from what this listing actually requires. Answered in your browser — nothing is sent to us, and we do not store your answers.

Do you have a Master's degree or higher (MSc, PhD, PharmD) in Computational Biology, Cheminformatics, Medicinal Chemistry, Biochemistry, or a closely related field, OR equivalent professional experience (e.g., 5+ years in computational drug discovery)?
Computational biology or cheminformatics domain expertise · required
Have you written and deployed production-quality Python code (not analysis scripts) for computational biology or cheminformatics workflows, with experience using version control (Git/GitHub)?
Production Python coding · required
Do you have hands-on experience with at least one major cheminformatics platform or toolkit (RDKit, KNIME, Schrödinger, OpenEye, MOE, or equivalent)?
Cheminformatics toolkit proficiency · required
Have you worked with public small-molecule or bioactivity databases (e.g., ChEMBL, PubChem, DrugBank) or internal assay/structure-activity datasets in a drug discovery context?
Small-molecule drug discovery datasets · preferred
Do you have experience building or deploying reproducible computational environments using Docker or similar containerization tools?
Docker or reproducible environments · preferred

0 of 5 answered

Included with this listing

Interview prep — 11 questions

Matched to what this platform's screening actually measures. Answer each out loud in 60–90 seconds: claim → concrete example → implication.

What it tests: Whether they've actually done SAR/QSAR or compound-target work at scale, and whether they can articulate the biological and chemical reasoning, not just run a script.
A strong answer: Names a specific dataset (ChEMBL, internal assay data, etc.), describes the computational method used (QSAR modeling, molecular docking, descriptor analysis), explains what they learned about the structure-activity relationship, and connects it to a real decision (compound prioritization, optimization hypothesis).
Where people lose it: Vague project description ('I did some cheminformatics analysis'), no mention of actual datasets or biological context, or describing output without explaining what the SAR/QSAR result meant for drug discovery.