BSCS @ Georgia Tech · Software Engineering Intern @ Travelers · Founder @ Signifi
Portfolio · Email · LinkedIn · Devpost
I build applied AI systems across cloud infrastructure, computer vision, robotics, and full-stack product development.
My favorite projects move beyond a model demo. They connect the model to reliable backend workflows, clear interfaces, evaluation, and real users.
- Building AI-assisted support workflows with AWS Bedrock and serverless infrastructure at Travelers
- Building Signifi, a requirements platform that turns raw notes into linked product specs, user stories, test cases, and interactive prototypes
- Leading Graption, a speaker-aware live-captioning platform, with a 15-person Georgia Tech team
- Researching multi-agent spacecraft autonomy with ROS2, MuJoCo, and decentralized learning
| Project | What it does | Highlights | Stack |
|---|---|---|---|
| Signifi | Turns notes and transcripts into connected requirements, engineering artifacts, and prototypes | Founder and lead developer; 3 pilot teams; $25K in startup cloud funding | TypeScript, Next.js, structured AI outputs, artifact graphs |
| Graption | Places live captions near the person speaking and recovers missed conversation context | Project lead; 0.74 F1 across 8 hours of audio-video; Hacklytics winner | React Native, FastAPI, WebSockets, OpenCV, MediaPipe, Whisper |
| Talking Fingers | Scores live sign-language practice from hand and body movement | Supports 37 reference signs, including dynamic J and Z motions | Swift, SwiftUI, Apple Vision, dynamic time warping |
| Gridly | Controls a physical smart plug and schedules device use for lower-carbon grid windows | Tested across two Georgia cities; Best Use of Vultr at EnergyHack | WebSockets, Vultr, PostgreSQL, IoT, Electricity Maps API |
| Veritas | Analyzes medical bills and coordinates email, research, and voice agents to support disputes | YHack 2026 K2 Think V2 Track finalist | Next.js, Supabase, Gemini, OpenAI, Perplexity, ElevenLabs, Twilio |
| Echolocate | Simulates how 25 AI agents respond to an economic policy inside an interactive city | Multi-agent memory, planning, opinion dynamics, and live analytics | LangGraph, FastAPI, Next.js, Phaser, WebSockets |
- AI / ML: AI agents, RAG, embeddings, LLM workflows, OpenCV, MediaPipe, Whisper, scikit-learn
- Backend / Cloud: AWS, Bedrock, Lambda, EventBridge, Step Functions, FastAPI, Spring Boot, PostgreSQL
- Frontend / Mobile: React, Next.js, TypeScript, React Native, SwiftUI
- Robotics / Systems: ROS2, MuJoCo, C++, Docker, Terraform, Linux
- Languages: Python, Java, C++, TypeScript, JavaScript, Swift, SQL
- 4-time hackathon winner with 2 additional track finalist finishes
- Bitcamp Best Moonshot Hack for Echolocate
- Hacklytics winner for Graption
- codeLinc winner for FinMate
- Best Use of Vultr at EnergyHack for Gridly
Summer 2027 software engineering internships in applied AI, backend and cloud infrastructure, or product engineering.
Email: akshajnadn@gmail.com
LinkedIn: linkedin.com/in/akshaj-nadimpalli
Portfolio: akshajnad.tech
Devpost: devpost.com/akshajnad