About
AI practitioner with hands-on experience training, evaluating, and quality-assuring LLM systems — scoring model outputs for relevance, coherence, and factual accuracy, annotating large datasets, and refining evaluation criteria to improve model performance.
Built and tested RAG pipelines and AI agents end-to-end, giving deep insight into how models reason, retrieve, and fail.
Exceptional attention to detail, strong critical reading and written communication, and a proven record of working independently in fully remote, high-ownership settings.
What I'm looking for
Data science roles,Ai/ml roles,ml engineer,ai trainer
Experience
AI/ML Engineer Intern
Jan 2026 – Mar 2026
• Evaluating outputs of LLM-powered RAG pipelines and AI agent systems in a high-impact R&D environment — reviewing generated responses against source documents for factual accuracy, relevance, and reasoning quality.
• Testing autonomous AI agents built with LangChain and MCP frameworks; documenting failure modes, scoring multi-step reasoning behaviour, and feeding structured findings back into prompt and retrieval improvements.
LangChainMCP Agent SystemsRAG Pipelines
DRDO (Defence Research & Development Organisation)
Lead Data Science Intern
Aug 2025 – Oct 2025
• Curated, cleaned, and categorized 50,000+ records to train BERT-based NLP models with TensorFlow and Scikit-learn — applying detailed labelling criteria consistently and validating data quality, reducing manual effort by 25%.
• Led a small remote team applying shared annotation guidelines across the dataset; reviewed edge cases, resolved labelling disagreements, and standardized criteria, driving a 15% increase in pipeline efficiency.
BERT Fine-tuningTensorFlowScikit-learn
Renu Sharma Foundation
Education
IINTM (IP University), Delhi
BCA · Computer Science
2023 – 2026
Certifications
Data Science Professional & Power BI Certification
IINTM
2024 – 2025
Cloud Computing & Software Testing
NPTEL
2023 – 2024
Skills
Power BIPyTorchTensorFlowScikit-learnChromaFAISSSQLFastAPINumPyPandasPythonBERT Fine-tuningGPT Fine-tuningMCP Agent SystemsRAG PipelinesLangChainFactuality reviewError analysisData labellingRubric-based scoring
Languages
English (Full professional proficiency)Hindi (Native or bilingual proficiency)
