HARSHAL SUDHAKAR PATIL

0 connections
About
Machine Learning Engineer with hands-on experience designing, containerizing, and deploying production-grade ML pipelines and agentic LLM workflows. Expert in building robust MLOps infrastructure, automated training pipelines, and real-time REST APIs using Python, Docker, CI/CD, and cloud platforms (AWS/Azure). Strong expertise in bridging data science and software engineering to scale high-throughput, containerized AI models.
Experience
AI/ML Engineer Intern

AiAdventure

November 2025 - June 2026

● Engineered and deployed a scalable production RAG pipeline to AWS and Azure via automated CI/CD workflows, integrating advanced MLOps governance frameworks including security guardrails, observability, and LLM gateways alongside performance evaluation metrics to ensure enterprise-grade reliability.

Data Science Intern

iNeuron Intelligence Pvt. Ltd.

May 2024 - November 2024

● Architected and deployed scalable financial models to cloud environments (AWS/Azure) using Docker and GitHub Actions CI/CD, while tracking model performance, data drift, and logging via MLflow, DVC, and DagsHub.

Education
Bachelor of Technology

G H Raisoni Institute of Engineering and Technology

August 2019 - August 2023

Field: Artificial Intelligence

Skills
Advanced
Generative AI & NLP Machine Learning & Deep Learning MLOps & Infrastructure Programming
Intermediate
Cloud & Engineering
Projects
Autonomous Data Analyst Agent

April 2026 - June 2026

Architected an autonomous multi-agent system using LangGraph to translate unstructured business queries into programmatic execution plans and compiled reports, integrated with multi-source ingestion pipelines and an interactive Streamlit dashboard featuring real-time graph tracking.

High-Performance PDF Knowledge Engine

February 2026 - April 2026

Engineered a production-grade RAG conversational QA system using LangChain and ChromaDB, Hybrid Search, Reranker, achieving 1.44s end-to-end latency with a 9-stage secured request lifecycle deployed via Docker, while integrating advanced security and observability layers such as prompt-injection scanning, exponential back-off retries, and real-time P95 latency tracking.

Contact Information
Quick Stats

0

Connections

0

Posts

2

Experiences