Lead AI Alignment & Evaluation Researcher

Wipro; San Francisco (CA); On-site; USD 240,000 - 375,000; Full time

Benefits offered by this job

Medical benefits

Dental benefits

Paid time off

Disability insurance

Job summary

Wipro is seeking a Lead Researcher for its AI Data Foundry in San Francisco to advance measurement and alignment for the next generation of LLMs and SLMs. You will architect preference data pipelines, design proactive benchmarks, and publish novel methodologies on arXiv and at top-tier conferences.

You will collaborate with domain researchers to synthesize ground-truth data into robust evaluation frameworks.

Qualifications

• Advanced degree in CS/AI/quantitative field.
• Expertise in RLHF/DPO/RLAIF and evaluation harnesses.
• Proficient in PyTorch, HuggingFace, and data manipulation.

Responsibilities

• Design alignment methodologies for human-in-the-loop and synthetic data.
• Build proactive, domain-specific benchmarks for advanced AI systems.
• Decontaminate data pipelines to detect overlaps and mitigate reward hacking.
• Publish research on evaluation frameworks and alignment science.
• Lead technical collaboration with researchers to produce scoreable evaluations.

Skills

RLHF/DPO/RLAIF experience

Bradley-Terry ranking

Data pipeline design

Benchmark design

Open research publication

Python

PyTorch

HuggingFace

Education

PhD or MSc in Computer Science / AI / Mathematics

Tools

PyTorch

HuggingFace

vLLM

EleutherAI LM Eval

HELM

Job description

Wipro is seeking a Lead Researcher for its AI Data Foundry in San Francisco to advance measurement and alignment for the next generation of LLMs and SLMs. You will architect preference data pipelines, design proactive benchmarks, and publish novel methodologies on arXiv and at top-tier conferences.

You will collaborate with domain researchers to synthesize ground-truth data into robust evaluation frameworks.