STEM Research Jobs in India

What You Will Do

• Plan and run controlled experiments
• Maintain reproducible research notebooks and clear deliverables
• Evaluate LLM outputs using prompt evaluation and QA evaluation
• Help define evaluation protocols and metrics
• Support RLHF-style evaluation workflows (preferred) including rubric design and structured human feedback
• Create/refine annotation rubrics for content safety labeling, NER, and task-specific guidelines compliance
• Analyze error patterns and recommend changes that improve robustness, safety, and generalization

Requirements

• Mid-Senior experience in applied research, data science, or engineering
• Strong experimental design and statistical reasoning
• Proficiency in Python and ML tooling; ability to interpret quantitative metrics and qualitative failure modes
• Familiarity with LLM evaluation concepts (prompt evaluation, rubric design, QA evaluation)
• Comfort working with labeling operations, annotation guidelines, and data quality processes

Preferred Qualifications

• Hands-on experience with RLHF evaluation, preference modeling, or human feedback programs
• Exposure to NLP and computer vision datasets, named entity recognition, and content safety labeling
• Experience building evaluation harnesses, automated QA checks, and documentation for repeatable execution