Janardhan Reddy Guntaka

AI Engineer & ML Systems Builder

Building production-grade LLM systems, RAG pipelines, and intelligent agents at scale

↓ Scroll to explore

About

I'm an AI engineer passionate about building production-grade systems at the intersection of large language models, machine learning, and cloud infrastructure. With expertise in LangChain/LangGraph agentic systems, RAG pipelines, and fine-tuning, I specialize in translating complex AI problems into scalable solutions.

Currently pursuing an M.S. in Data Science at University at Buffalo, SUNY, with prior industry experience at AUDREE LLC working on Python microservices, AWS infrastructure, and CI/CD pipelines.

I'm drawn to challenging problems in agentic AI, HIPAA-compliant NLP, demand forecasting systems, and computer vision. My work spans LLM engineering, MLOps, and full-stack development.

Specialization
LLM-powered agents, RAG pipelines, fine-tuning, and production ML systems
Focus Areas
Agentic AI, HIPAA compliance, NLP, time-series forecasting, computer vision
Cloud & Infra
AWS (ECS Fargate, Lambda, S3), Docker, GitHub Actions, MLOps

Skills & Technologies

LLM & AI Engineering

LangChain LangGraph RAG Pipelines OpenAI API Prompt Engineering Gemini AI

Vector Databases & Search

pgvector ChromaDB Vector Search Semantic Retrieval Embeddings

Programming Languages

Python TypeScript JavaScript Java SQL

Web & Backend

FastAPI React Next.js Node.js Spring Boot Streamlit

ML & Data Engineering

PyTorch HuggingFace Fine-tuning LightGBM Scikit-learn MLflow Hopsworks

Cloud & DevOps

AWS ECS Fargate AWS Lambda S3 DynamoDB Docker GitHub Actions Kafka RabbitMQ

Databases

PostgreSQL MongoDB Supabase DynamoDB

Other Tools & Platforms

Keycloak OAuth2 HIPAA Compliance Git Linux

Featured Projects

TariffIQ

Production agentic RAG platform for AI-powered HTS tariff classification and duty calculation with a 7-stage LangGraph pipeline deployed on AWS ECS Fargate.

Python FastAPI LangGraph OpenAI pgvector Next.js AWS ECS
View on GitHub

Aegis Health

Clinical NLP platform with fine-tuned Bio_ClinicalBERT (94.99% recall) for real-time PHI detection and HIPAA governance over LLM responses.

Python FastAPI HuggingFace PyTorch Bio_ClinicalBERT React
View on GitHub

NYC Taxi Demand Forecasting

End-to-end production ML system processing millions of trip records through 3 automated GitHub Actions pipelines with MLflow tracking.

Python LightGBM Hopsworks MLflow GitHub Actions Streamlit
View on GitHub

Citi Bike Demand Forecasting

Production ML system with hourly automated inference pipelines and real-time Streamlit dashboard for bike demand prediction.

Python LightGBM Hopsworks MLflow GitHub Actions Streamlit
View on GitHub

AI Fitness Tracker

5-service Java microservices platform with Keycloak OAuth2, RabbitMQ async messaging, and Gemini AI workout recommendations.

Java Spring Boot RabbitMQ Keycloak PostgreSQL MongoDB Docker
View on GitHub

CLIP Chest X-Ray

Fine-tuned OpenAI CLIP on 7,430 chest X-ray image-report pairs, improving Top-1 retrieval accuracy by 41% over zero-shot baseline.

Python PyTorch CLIP ViT-B/32 HuggingFace Scikit-learn
View on GitHub

Experience

Software Engineer Intern

AUDREE LLC
May 2023 – July 2024
Designed and implemented Python microservices for event-driven architecture. Worked with AWS Lambda for serverless computing, Kafka for real-time data streaming, and DynamoDB for NoSQL data persistence. Built and maintained CI/CD pipelines using GitHub Actions, containerized applications with Docker, and collaborated on cloud infrastructure optimization.

Education & Certifications

M.S. Data Science

University at Buffalo, SUNY
2024 – 2025

B.S. Computer Science

Koneru Lakshmaiah University
2020 – 2024

Professional Certifications

AWS Certified Solutions Architect – Associate
AWS Certified Cloud Practitioner

Let's Connect

I'm always interested in AI engineering opportunities, collaboration, and discussing innovative projects.