👋 Hello, I'm

Gayathri Chilukala

|

Passionate about building intelligent systems that solve real-world problems. Currently architecting LLM backend services at GEICO and exploring the frontiers of Generative AI.

View My Work Get in Touch
Gayathri Chilukala
Scroll Down

01. Education

Florida Atlantic University

M.S. in Computer Science (AI Specialization) · Aug 2023 — Dec 2024 · Boca Raton, FL
  • GPA: 3.96/4.0; Gangal Family Endowed Graduate Scholarship (Top 1% of CS students).
  • Focus: AI/ML, Deep Learning, NLP; LLMs, Generative AI, and multi‑agent systems.
  • Advanced coursework: AI algorithms, neural networks, computer vision, distributed systems.

B.E. / B.Tech — Computer Science & Engineering

India · 2019 — 2023
  • Strong foundations in data structures, algorithms, databases, and full‑stack development.
  • Multiple projects demonstrating end‑to‑end engineering skills.

Academic Highlights

  • Graduate Teaching Assistant: taught 300+ students AI/ML & C++, improving outcomes 40–50%.
  • Hackathon winner: Caption Creator Pro (sub‑2.1s captions with SambaNova Llama models).
3.96FAU GPA
300+Students Taught
Top 1%Scholarship

02. Work Experience

Software Engineer – AI Solutions @ GEICO

Chevy Chase, MD · Jan 2025 – Present
  • Architected & deployed production LLM backends with Python + FastAPI on Azure (vLLM), enabling scalable AI features.
  • Improved claims model accuracy by ~2.5% and cut latency ~50% via prompt tuning and LLaMA evaluation; reduced costs.
  • Built a Generative AI platform for document processing (PDF/CSV/DOCX/TXT) with role‑based access & dynamic UI.
  • Created an LLM eval pipeline (30% faster via parallelization) and a Snowflake dashboard for model monitoring.
  • Implemented robust token counting across 8 models with automated guardrails and error handling.
  • Designed an event‑driven logging service (Kafka → Postgres) for auditing and real‑time monitoring.
  • Shipped an MCP tools marketplace to register, discover, and test Model Context Protocol tools.

Graduate Teaching Assistant – AI & C++ @ FAU

Boca Raton, FL · Jan 2024 – Dec 2024
  • Developed and taught advanced AI/ML algorithms including gradient descent, ridge regression, and coordinate descent
  • Created comprehensive hyperparameter tuning frameworks using cross-validation techniques (k-fold, LASSO, ridge regression)
  • Improved student learning outcomes by 40-50% through innovative ML performance visualization tools built with ggplot2
  • Instructed 300+ students in C++ object-oriented programming and algorithm optimization

03. Skills & Tech Stack

AI/ML Frameworks

TensorFlow

PyTorch

Keras

Hugging Face

LangChain

OpenAI API

scikit-learn

NLTK

Backend Development

Python

FastAPI

Django

Node.js

Java

JavaScript

C++

Go

C#

PHP

TypeScript

Frontend & FullStack

React.js

Next.js

Express.js

React Native

Cloud & DevOps

Azure

AWS

GCP

Docker

Kubernetes

Git

Kafka

CI/CD Pipelines

Databases & Analytics

PostgreSQL

MySQL

MongoDB

Firestore

Snowflake

AI Specializations

Large Language Models (LLMs)

RAG (Retrieval Augmented Generation)

Generative AI

Natural Language Processing

Computer Vision

Agentic AI Systems

03. Featured Projects

Pic2Plot - Multi-Agent AI System

Featured Project

Sophisticated multi-agent system with 4 specialized modules using GPT-4.1, LLaMA 3 Vision, and DALL-E 3. Achieved 40% cost reduction through adaptive optimization and intelligent module coordination for computer vision tasks and image processing workflows.

  • GPT-4.1
  • LLaMA 3 Vision
  • DALL-E 3
  • Multi-Agent AI

Caption Creator Pro - Hugging Face Winner 🏆

Featured Project

Hugging Face Hackathon Winner! High-performance multimodal AI platform with SambaNova Llama models achieving sub-2.1 second caption generation. Integrated advanced image processing with state-of-the-art language models for real-time intelligent caption generation.

  • Hugging Face
  • LLaMA
  • SambaNova
  • Multimodal AI

SafetyMapper Platform

Featured Project

Real-time incident reporting platform with AI-powered safety recommendations using Google Gemini AI. Features intelligent route analysis system for transportation safety with Flask backend deployed on Google Cloud Platform for scalable incident tracking and prediction.

  • Flask
  • GCP
  • Gemini AI
  • Route Analysis
View More on GitHub

05. Get In Touch

Let's Work Together

I'm passionate about AI/ML innovation and always excited to discuss cutting-edge projects, research collaborations, or AI solutions. Whether you have questions about Generative AI, LLMs, or want to explore potential collaborations, let's connect!

Email

chilukalagayathri@gmail.com

Location

Washington, D.C., USA

Response Time

Within 24 hours

Connect With Me