AI/ML Engineer — AI Governance Audit Platform

Job Title: AI/ML Engineer — AI Governance Audit Platform
Experience: 2-5 Years
Employment Type: Full-Time
Location : Gurugram, Haryana (Onsite Role)

About the role

We're building an AI governance audit SaaS that scans a company's AI tool stack against regulatory and custom guidelines (EU AI Act, NIST AI RMF, ISO 42001, and uploaded policy documents) and generates a compliance gap report. We're looking for an AI/ML Engineer to build the core intelligence layer: extracting checkable requirements from guideline documents and matching them against a company's AI tool inventory.

What you'll work on

• Build an LLM-based pipeline to parse uploaded guideline documents (PDF/Word) and extract structured, checkable requirements
• Design and maintain the rules engine that maps requirements from built-in frameworks (EU AI Act, NIST AI RMF, ISO 42001) to risk tiers and controls
• Build the gap-analysis/matching logic that scores a company's AI tool stack against extracted requirements
• Work on prompt design, retrieval, and evaluation to keep extraction accuracy high and hallucination low
• Define confidence scoring so low-certainty extractions get flagged for human review rather than silently entering the report
• Collaborate with backend engineers to expose this as an API consumed by the product
• Continuously evaluate and improve accuracy as we add more frameworks and guideline formats

What we're looking for

• 2 to 5 years of experience building production ML/NLP systems, ideally with LLMs
• Hands-on experience with LLM APIs (OpenAI, Anthropic, or similar), prompt engineering, and structured output extraction
• Experience with document parsing (PDF/Word) and unstructured-to-structured data pipelines
• Comfortable with Python, and familiar with vector databases/embeddings for retrieval-based matching
• Understanding of evaluation methods for LLM outputs (accuracy, hallucination detection, human-in-the-loop review flows)
• Bonus: prior exposure to compliance, GRC, legal-tech, or regulatory domains
• Bonus: experience building systems that need to cite sources/evidence for their outputs (important for audit-ready reports)

Nice to have

• Familiarity with EU AI Act, NIST AI RMF, ISO 42001, or similar frameworks
• Experience in an early-stage startup or 0-to-1 product environment.

Pay: ₹600,000.00 - ₹800,000.00 per year

Benefits:

• Provident Fund

Work Location: In person