Chandrani Ganguly Roy

About Chandrani Ganguly Roy

I am a Trust & Safety and Data Analytics professional with 6+ years of experience across Google and Meta, focused on abuse detection, policy enforcement, privacy compliance, and AI/ML data quality. I specialize in investigating large‑scale abuse networks, designing scalable enforcement workflows, and ensuring that high‑stakes content and user data remain safe and compliant.At Google, I investigated spam and fraud across global developer ecosystems, using SQL and pattern analysis to uncover coordinated abuse, refine detection heuristics, and support high‑impact enforcement decisions on Google Play. I also led privacy and permissions reviews, ensuring mobile apps adhered to data minimization and family‑friendly content standards.At Meta, I support AI/ML development as an AI Quality Auditor and Data Labeling Analyst, performing high‑precision labeling and QA across text, image, and video datasets used to train and evaluate LLMs. I maintain 90%+ QA accuracy, help refine annotation schemas and guidelines, and have reduced rework and labeling defects by 35% through structured feedback and error trend analysis.

Experience

1. Ai Quality Auditor (Ai Ml)

   Meta

   May 2025 — Present · US

   Driving high-quality data pipelines for LLM training and evaluation through advanced annotation QA, guideline refinement, and cross-functional collaboration.Audit large-scale text, image, and video datasets supporting LLM training and evaluation workflows, ensuring high precision and policy complianceConsistently maintain 95%+ QA accuracy, improving overall data quality and model input reliabilityReduced annotation rework and defects by 35% by identifying recurring error patterns and implementing structured feedback loopsProvide targeted coaching and calibration support to annotators, improving consistency across nuanced and culturally sensitive content categoriesAuthored 12+ guideline updates, clarifying ambiguous edge cases and strengthening evaluation standards across workflowsPartner with product and engineering teams to refine annotation schema, improve tooling, and enhance QA productivityCollaborate with internal and external vendor teams, ensuring alignment on quality benchmarks and evaluation standardsAnalyze systemic inconsistencies and edge cases to proactively mitigate quality risks in AI pipelinesContribute to improving human-in-the-loop evaluation frameworks for scalable and reliable model development

Education

• Bangalore University

  Bachelors of Computer Science

Skills

• Sourcing
• Vendor Management
• Recruiting
• Business Analysis
• Team Management
• Management
• Human Resources
• Benefits Negotiation
• Crm
• Talent Acquisition
• Screening Resumes
• Contract Recruitment
• IT Recruitment
• Staff Augmentation
• Software Development Life Cycle (Sdlc)
• Business Development
• Internet Recruiting
• Screening
• Technical Recruiting