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

Architected an enterprise RAG assistant using LangChain and FAISS to optimize sub-300ms retrieval for high-stakes search. • Engineered agentic reasoning logic via MMR and prompt engineering to mitigate hallucinations and achieve 97% accuracy. • Developed a modular ETL pipeline for automated ingestion to ensure observability and 1:1 mapping with verified source pages. • Translated complex business requirements into technical system design, delivering a production-grade interface via Streamlit.…Architected an enterprise RAG assistant using LangChain and FAISS to optimize sub-300ms retrieval for high-stakes search. • Engineered agentic reasoning logic via MMR and prompt engineering to mitigate hallucinations and achieve 97% accuracy. • Developed a modular ETL pipeline for automated ingestion to ensure observability and 1:1 mapping with verified source pages. • Translated complex business requirements into technical system design, delivering a production-grade interface via Streamlit.

Ashwin Sonawane

AI Engineer, Full Stack Developer, Web Developer, +2





Architected an enterprise RAG assistant using LangChain and FAISS to optimize sub-300ms retrieval for high-stakes search. • Engineered agentic reasoning logic via MMR and prompt engineering to mitigate hallucinations and achieve 97% accuracy. • Developed a modular ETL pipeline for automated ingestion to ensure observability and 1:1 mapping with verified source pages. • Translated complex business requirements into technical system design, delivering a production-grade interface via Streamlit.…Architected an enterprise RAG assistant using LangChain and FAISS to optimize sub-300ms retrieval for high-stakes search. • Engineered agentic reasoning logic via MMR and prompt engineering to mitigate hallucinations and achieve 97% accuracy. • Developed a modular ETL pipeline for automated ingestion to ensure observability and 1:1 mapping with verified source pages. • Translated complex business requirements into technical system design, delivering a production-grade interface via Streamlit.

Available to hire

Architected an enterprise RAG assistant using LangChain and FAISS to optimize sub-300ms retrieval for high-stakes search.
• Engineered agentic reasoning logic via MMR and prompt engineering to mitigate hallucinations and achieve 97% accuracy.
• Developed a modular ETL pipeline for automated ingestion to ensure observability and 1:1 mapping with verified source pages.
• Translated complex business requirements into technical system design, delivering a production-grade interface via Streamlit.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Work Experience

AI Engineer at Community Dreams Foundation

May 1, 2025 - Present

Architected an enterprise RAG assistant using LangChain and FAISS to optimize sub-300ms retrieval for high-stakes search. Engineered agentic reasoning logic via MMR and prompt engineering to mitigate hallucinations and achieve 97% accuracy. Developed a modular ETL pipeline for automated ingestion, ensuring observability and 1:1 mapping with verified source pages. Delivered a production-grade interface via Streamlit.

Software Developer at Goupbrod

July 1, 2022 - June 1, 2023

Launched a recommendation engine that matched 50K+ student profiles to programs, improving precision by 30%. Designed a multilingual real-time ETL pipeline for profile validation and transformation, reducing onboarding time by days. Built full-stack dashboards to visualize engagement and model performance, helping stakeholders boost student conversions.