Summary
Overview
Work History
Education
Skills
Awards
Projects
Languages
Timeline
Generic

Devi Priya Sugrivagari

Summary

AI/Cloud Software Engineer | Full-Stack Development | LLM & Cloud Infrastructure

Overview

3
3
years of professional experience

Work History

Software Engineering Intern - AI

EyCrowd, Inc.
San Francisco
03.2025 - 05.2025
  • Developed AI-driven chatbots using GPT-4 and LangChain and React Native apps with LLM integration, automating 70% of customer workflows and reducing API calls by 40% via memoization while ensuring cross-platform consistency.
  • Deployed auto-scaling AWS infrastructure (ECS/Docker) handling 1K+ RPS at 99.9% uptime for real-time AI inference.

Graduate Student Assistant

Cumberland University
Lebanon
08.2024 - 12.2024
  • Built an AI-powered automated grading system using LangChain and sentence-transformers, leveraging semantic similarity scoring to reduce grading time by 90% and lower false positives by 35% compared to regex-based methods.
  • Optimized Docker-based microservices (60% smaller deployment images) and load balancing, improving API throughput by 45%.
  • Architected a high-concurrency backend (500+ users) using asynchronous processing and resource optimization, achieving sub-2-second response times for real-time AI evaluation tasks.

Full Stack Engineer

Webdizer Software Solutions
Hyderabad
09.2022 - 11.2023
  • Worked in Agile (scrum) environment with 15-day sprint cycles.
  • Built MVC-based web applications using JSP, Angular, Vue.js, JavaScript, HTML5, CSS3, AJAX.
  • Developed enterprise J2EE web applications using Spring MVC, Hibernate with UI in ReactJS, C#, HTML5, CSS3, TypeScript.
  • Optimized database queries across MySQL, PostgreSQL, and SQL Server, improving reliability by 30%.

Education

Master of Science - Information Technology Project Management

Cumberland University
12.2025

Bachelor of Technology - Information Technology

SVCE
08.2022

Skills

  • Python
  • Java
  • C
  • C
  • SQL
  • JavaScript
  • TypeScript
  • HTML
  • CSS
  • Shell scripting
  • Linux
  • Transformer-based LLMs
  • LangChain
  • HuggingFace
  • PyTorch
  • TensorFlow
  • BERT
  • Semantic Search
  • OpenCV
  • NLTK
  • AWS (Lambda, S3, EC2)
  • GCP
  • Azure
  • Docker
  • Kubernetes
  • CI/CD (GitHub Actions, Jenkins)
  • Git
  • Jupyter
  • React
  • Nodejs
  • Threejs
  • Flask
  • MongoDB
  • MySQL
  • PostgreSQL
  • REST
  • Microservices
  • Postman
  • JIRA
  • Figma
  • React Native
  • Cursor
  • Websocket
  • HTTP

Awards

  • CrewAI: Multi-Agent Systems Development, Comprehensive training covering task automation and advanced agent architectures
  • DeepLearning.AI & AWS: Generative AI with Large Language Models
  • Hackathon Winner, SoDAhacks - Assess.AI: Smarter Assessment for Smarter Learning, 10/01/24

Projects

  • GenAI-Powered Job Matching Engine, AWS (EC2, S3), Streamlit, web scraping, REST APIs, Pandas, NLTK, Developed a GenAI job-matching engine using BERT embeddings and LangChain, automating resume-to-job compatibility scoring (0-100%) via cosine similarity (scikit-learn), with AWS-hosted Streamlit UI, customizable keyword filters, and multi-platform job scraping.
  • Serverless Video Processing Pipeline with Hybrid Cloud Integration, Infrastructure as Code (IaC), container orchestration, auto-scaling, Engineered a serverless video processing pipeline using AWS Lambda, OpenFaaS, and Kubernetes (EKS), achieving 98% cost efficiency vs. on-prem solutions. Orchestrated hybrid cloud workflows for event-driven media workflows.
  • Assess.AI - AI-Powered Academic Grading System, Llama-3.1-70B, Sentence-BERT, semantic search, PyMuPDF, NLTK, Built an LLM-driven grading platform using Llama-3.1-70B and semantic search (ChromaDB + Sentence-BERT), automating 70% of manual evaluations with 92% accuracy in feedback alignment and 30% fewer plagiarism false positives.
  • Multi-Objective Autonomous Agents - LMDP Framework, Reinforcement Learning, OpenAI Gym, Warehouse Navigation, Engineered safety-first autonomous agents via lexicographic MDPs (PyTorch), reducing safety violations by 40% and achieving 85% task success in stochastic warehouse navigation.

Languages

  • Python
  • Java
  • C++
  • C
  • SQL
  • JavaScript
  • TypeScript
  • HTML
  • CSS
  • Shell scripting
  • Linux

Timeline

Software Engineering Intern - AI

EyCrowd, Inc.
03.2025 - 05.2025

Graduate Student Assistant

Cumberland University
08.2024 - 12.2024

Full Stack Engineer

Webdizer Software Solutions
09.2022 - 11.2023

Master of Science - Information Technology Project Management

Cumberland University

Bachelor of Technology - Information Technology

SVCE
Devi Priya Sugrivagari