Hire this AI Trainer
Sign in or create an account to invite AI Trainers to your job.
Google Cloud Vertex AI
OneForma
Text
Image
Video
Entity (NER) Classification
Red Teaming
Fine-tuning
RLHF
Evaluation/Rating
Bounding BoxI am an AI Engineer and published researcher with over 4 years of experience building production-grade machine learning systems and leading responsible AI initiatives. My core expertise lies at the intersection of advanced LLM evaluation, red-teaming, and model robustness. I have a proven track record of designing adversarial frameworks to identify vulnerabilities across frontier models (including GPT-5 and Gemini 3 Pro) , as well as developing advanced fine-tuning methodologies to optimize multilingual PII extraction and data integrity. With a strong technical foundation combining an integrated Master's/Bachelor's from IIT Delhi and extensive hands-on experience across GCP, Azure, and PyTorch, I excel at breaking down complex, high-level AI behaviors into precise, measurable data. What sets me apart for high-tier AI training data initiatives is my dual capability in both applied research and practical engineering. Having authored first-author research on scalable annotation frameworks for responsible AI , I deeply understand the mathematical and logical structures required to evaluate model reasoning, mitigate algorithmic bias , and design strict, high-quality data pipelines. I am uniquely equipped to tackle expert-level tasks that require rigorous logic, adversarial auditing, code evaluation, and deep domain knowledge in machine learning architectures. Education includes Bachelor of Technology, Indian Institute of Technology (IIT), Delhi (2020) and Master of Technology, Indian Institute of Technology (IIT), Delhi (2020). AI-training focus includes data types such as Text and labeling workflows including Fine-tuning, Evaluation, and Rating.
TextRed TeamingConducted rigorous safety testing and vulnerability mapping on frontier large language models to eliminate security and alignment vulnerabilities. Authored and evaluated thousands of complex, multi-agent adversarial prompts using advanced frameworks to probe for jailbreaks, tracking exploit pathways, and labeling model outputs for explicit policy compliance and risk mitigation.
TextEntity (NER) ClassificationDeveloped and managed a high-complexity annotation framework across a dataset of 40,000 text samples to train large language models to securely identify and extract sensitive data across 13 global locales. Established a rigorous multi-tier quality pipeline where initial annotator outputs were validated via human-in-the-loop expert auditing, successfully achieving a high-quality benchmark of 95% or above recall and a 5% or less False Positive Rate (FPR). Utilized this ultra-clean data to evaluate model robustness and optimize multilingual token-classification pipelines.
Bachelor of Technology, Biochemical Engineering and Biotechnology
Master of Technology, Biochemical Engineering and Biotechnology
AI Engineer
Data Science Analyst
