Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

AI/ML professional specializing in LLM evaluation, RAG optimization, and retrieval pipelines. I build Python-based benchmarking and REST API systems to improve evaluation consistency, dataset quality, and response grounding across multimodal AI workflows. I also design scalable data/ETL pipelines and backend services using FastAPI/Django, SQL, and vector databases, with strong focus on performance, testing, CI/CD, and production observability.…AI/ML professional specializing in LLM evaluation, RAG optimization, and retrieval pipelines. I build Python-based benchmarking and REST API systems to improve evaluation consistency, dataset quality, and response grounding across multimodal AI workflows. I also design scalable data/ETL pipelines and backend services using FastAPI/Django, SQL, and vector databases, with strong focus on performance, testing, CI/CD, and production observability.

Dayamay Das

AI Engineer, Data Scientist, Data Engineer, +2





AI/ML professional specializing in LLM evaluation, RAG optimization, and retrieval pipelines. I build Python-based benchmarking and REST API systems to improve evaluation consistency, dataset quality, and response grounding across multimodal AI workflows. I also design scalable data/ETL pipelines and backend services using FastAPI/Django, SQL, and vector databases, with strong focus on performance, testing, CI/CD, and production observability.…AI/ML professional specializing in LLM evaluation, RAG optimization, and retrieval pipelines. I build Python-based benchmarking and REST API systems to improve evaluation consistency, dataset quality, and response grounding across multimodal AI workflows. I also design scalable data/ETL pipelines and backend services using FastAPI/Django, SQL, and vector databases, with strong focus on performance, testing, CI/CD, and production observability.

Available to hire

AI/ML professional specializing in LLM evaluation, RAG optimization, and retrieval pipelines. I build Python-based benchmarking and REST API systems to improve evaluation consistency, dataset quality, and response grounding across multimodal AI workflows.

I also design scalable data/ETL pipelines and backend services using FastAPI/Django, SQL, and vector databases, with strong focus on performance, testing, CI/CD, and production observability.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Language

Work Experience

Deccan AIAI Expert at Deccan AIAI

June 1, 2026 - Present

Developed Python-based evaluation pipelines and REST API workflows to automate LLM benchmarking, improving evaluation consistency and reducing manual analysis across multimodal AI systems. Built and evaluated Retrieval-Augmented Generation (RAG) workflows using vector embeddings, semantic search, FAISS/Pinecone vector databases, document chunking, reranking strategies, and retrieval optimization to improve contextual grounding and response quality. Engineered datasets for LLM training by running 500+ prompt engineering, response ranking, preference optimization, and Supervised Fine-Tuning (SFT) tasks, achieving 98%+ annotation accuracy. Technologies include Python, SQL, REST APIs, LangChain, FAISS, Pinecone, RAG, and LLM evaluation.

Data Science and Analytics Intern at Zidio Development

September 1, 2025 - December 1, 2025

Built Python-based ETL pipelines for AI/ML datasets handling 10,000+ daily records while improving data quality and accelerating preprocessing by 30%. Optimized SQL queries and JSON parsing to reduce execution time by 40%, improving data processing performance for downstream tasks and ensuring high-quality deliverables. Collaborated with cross-functional teams to improve backend services, contributed to tooling enhancements, and supported a 20% increase in concurrent query volume. Technologies include Python, Excel, SQL, Machine Learning, AWS (EC2, S3), Deep Learning, NLP, LLM/LangChain, and vector database usage.