Bioinformatics Benchmark Designer - Single-Cell Genomics

Obsidian; New York (NY); Vor Ort; USD 83.000 - 124.000; Teilzeit

Zusammenfassung

Obsidian is seeking a Bioinformatics & Computational Single-Cell Genomics Expert to design graduate-level problems for AI systems. You will craft tasks that use advanced scientific software to run simulations, interpret results, and design experiments.

You will test problems against state-of-the-art AI models, iterating until the target difficulty is met, with a focus on cell-type annotation, pseudotime, and multi-omic integration using Scanpy, scVelo, Squidpy, and Gudhi.

Qualifikationen

• Graduate-level training in a relevant STEM field (MS, PhD, or equivalent research experience).
• Proven proficiency with at least one of the listed scientific software libraries, shown through publications, open-source contributions, or professional work.
• Strong Python skills for writing problem setups, oracle functions, and solution validators.
• Ability to work independently and refine problem designs based on feedback.
• Comfortable in a Linux/terminal environment with remote compute sandboxes.
• Available for at least 15–20 hours per week.

Aufgaben

• Design problems requiring skilled use of specialized scientific software.
• Test problems against AI models and refine until the target difficulty is reached.
• Create problems around cell-type annotation, pseudotime ordering, or multi-omic integration.
• Ensure problems require strategic planning and information extraction beyond straightforward computation.

Kenntnisse

Python

Problem design

Independent work

Ausbildung

MS or PhD in STEM or equivalent

Tools

scanpy

scvelo

squidpy

gudhi

Jobbeschreibung

Obsidian is seeking a Bioinformatics & Computational Single-Cell Genomics Expert to design graduate-level problems for AI systems. You will craft tasks that use advanced scientific software to run simulations, interpret results, and design experiments.

You will test problems against state-of-the-art AI models, iterating until the target difficulty is met, with a focus on cell-type annotation, pseudotime, and multi-omic integration using Scanpy, scVelo, Squidpy, and Gudhi.