Open to opportunities

Keerthi Shivapur

@keerthishivapur

Message

Generative AI & ML developer building reliable RAG and embedding retrieval systems.

Zimbabwe

Message

What I'm looking for

I want to build and evaluate real-world Generative AI systems—RAG with embeddings—using rigorous prompt testing, annotation, and hallucination detection to improve accuracy, relevance, and reliability.

I’m a Generative AI and Machine Learning enthusiast with hands-on experience building embedding-based retrieval systems and Retrieval-Augmented Generation (RAG) applications. I focus on prompt engineering, vector search, and making LLM outputs dependable through evaluation.

In my work, I develop AI applications using Python, FAISS, and transformer-based models. I’m especially interested in improving AI systems through structured testing, annotation, and response evaluation, with an emphasis on accuracy, relevance, and hallucination detection.

I’ve built projects like a cross-modal audio retrieval system (text-to-audio search) and a multi-source RAG assistant over PDFs, CSVs, and FAQs. I add confidence scoring and self-correction triggers to detect low-confidence or hallucinated responses, grounding answers in retrieved context to strengthen real-world reliability.

Experience

Work history, roles, and key accomplishments

Current

AI & ML Cloud Intern

Current

SuperMentor

Worked on AI/ML workflows including model evaluation, debugging, and performance analysis. Supported mentor-based evaluation processes and strengthened Python coding, logic, and analytical thinking through documentation and troubleshooting.

Python Machine Learning Model Evaluation Debugging Performance Analysis Technical Documentation Analytical Thinking Evaluation Pipelines