Open to opportunities

Saivedant Hava

@saivedanthava

Message

I build production AI systems, agentic RAG platforms, and high-throughput LLM infrastructure.

United States

Message

What I'm looking for

I'm looking to build production AI and LLM systems where I can improve reliability, retrieval quality, model governance, and scalable cloud deployment.

I've built production AI systems at AIA Enterprises and QuantifAI, including an agentic RAG platform that increased query resolution by 58% and semantic search across 80K+ records that reduced document lookup to under 90 seconds.

At QuantifAI, I trained LightGBM models on 278K rows and developed an end-to-end MLOps pipeline covering ingestion, feature engineering, training, risk allocation, explainability, backtesting, and reproducible inference. I also shipped advisor-facing Streamlit dashboards and a Next.js portfolio optimizer for 100–1000 portfolio suites.

I build reliable AI infrastructure in Python and Go, from an open-source LLM gateway sustaining 58K req/s to privacy-first local code review systems. I also contribute fixes and improvements to projects including Optuna, Ray Tune, Firecrawl, Kubeflow Katib, BentoML, OpenLIT, and OpenLLMetry.

Experience

Work history, roles, and key accomplishments

Current

AI Engineer

Current

AIA Enterprises LLC

Aug 2025 - Present (1 year)

Architected a production agentic RAG system with Ollama, Qdrant, and FastAPI, lifting query resolution by 58%. Built a semantic search pipeline over 80K+ records on AWS, cutting document lookup to under 90 seconds.

Python fastAPI Ollama Qdrant Docker Docker Compose AWS S3 Lambda RDS Embeddings Semantic Search RAG Agentic AI Prompt Engineering GitHub Actions Terraform CI CD

Current

AI Engineer

Current

QuantifAI

May 2025 - Present (1 year 3 months)

Trained LightGBM models on 278K rows with purged time-series cross-validation, achieving test IC of 0.0338 and PBO of 0.0. Designed an end-to-end MLOps pipeline and shipped advisor-facing dashboards.

Python LightGBM Time Series Analysis Cross Validation MLOps MLFlow Streamlit Next.js TypeScript SHAP Model Evaluation Feature Engineering Docker GitHub Actions

Education

Degrees, certifications, and relevant coursework

University of Dayton

Master of Science, Computer Science

2023 - 2025

Pursued a Master of Science in Computer Science with coursework in Artificial Intelligence, Natural Language Processing, Software Engineering, Computer Vision, and Data Visualization.