Evaluation Research2 openings
Research Scientist
Create rigorous evaluation methods, benchmark designs, and research-backed measurement systems for frontier AI workflows.
Remote-first, IST/CET overlap
4+ years in AI research, evaluation science, or applied ML research
Review Deadline: 3 Nov 2026
KEY RESPONSIBILITIES
- Design studies, metrics, benchmark tasks, and validation plans.
- Analyze model behavior, reviewer agreement, and benchmark drift.
- Publish internal research notes and client-facing technical findings.
QUALIFICATIONS & PREREQUISITES
Research record in ML, NLP, agents, benchmarking, human evaluation, or safety.
RELEVANT SKILLS & TOOLS
Research designBenchmarkingNLPStatistical analysisTechnical writing
SUBMIT APPLICATION
Apply for Research Scientist
Reviewed directly by senior practitioners on the Evaluation Research team. No automated rejection filters.
Prefer email? Send your resume to careers@evalixa.comDon't see your specific role?
We are always keen to speak with exceptional systems engineers, red-teamers, and credentialed domain fellows. If you build with rigor, write to us directly.
✓ 100% Practitioner Review✓ Response in < 5 Days✓ Remote-First Worldwide