Sivaji Alla

Contributed to ML platform development for LLM workflows by building Python services and data pipelines for model development, evaluation, and AI application workflows. Developed dataset preparation, validation, experiment tracking, and model-output analysis workflows to enable repeatable evaluation across AI development cycles. Built retrieval components for RAG workflows using embeddings and vector search, evaluating retrieval precision and response groundedness across 1,000+ evaluation queries. Maintained platform components via Git-based development, automated testing, code reviews, and CI/CD workflows, and collaborated with stakeholders on multiple AI platform initiatives to support production readiness.

Contributed to the ML platform supporting LLM workflows by building Python-based services and data pipelines for model development, evaluation, and AI application workflows. Developed dataset preparation/validation and evaluation workflows with experiment tracking and output analysis. Built RAG retrieval components using embeddings, vector search, and LLM integration, evaluating retrieval precision, response relevance, and groundedness across 1,000+ evaluation queries. Maintained platform components with Git-based development, automated testing, code reviews, and CI/CD workflows, collaborating on multiple AI platform initiatives.

Contribute to ML platform development supporting LLM workflows by building Python services and data pipelines for model development, evaluation, and AI application workflows. Developed repeatable dataset preparation and evaluation workflows (validation, experiment tracking, output analysis). Built RAG retrieval components using embeddings and vector search integrated with LLMs, evaluating retrieval precision, response relevance, and groundedness across 1,000+ evaluation queries. Developed and maintained platform components using Git, automated testing, code reviews, and CI/CD workflows while collaborating on 5+ AI platform initiatives.