resume

Dhairya Shah

Languages: TypeScript, JavaScript (Node.js), Rust, SQL, Python, C++
Backend: Fastify, FastAPI, Express, PostgreSQL, MongoDB, GraphQL, WebSockets
DevOps & Cloud: Docker, AWS Bedrock, Prometheus, Grafana, Loki, Serverless Functions, GitHub Actions

USThing (All in One App)

Architected and maintained backend services for 14,000+ active student users, handling high-concurrency traffic.
Engineered a Real-Time Chat System using PostgreSQL and Fastify over WebSockets, reducing message latency by 40% compared to polling methods.
Implemented production-grade observability pipelines using Prometheus and Loki, enabling proactive error detection and reducing downtime.
Managed backend task prioritization and CI/CD workflows, ensuring smooth deployment cycles for the app team.

Rice Robotics

Optimized inter-process communication (IPC) protocols for autonomous robots, improving command execution speed.
Collaborated with hardware teams to debug complex electromechanical integration issues in a production environment.

M-Labs Limited

Developed compiler tooling in Rust to translate Python kernels into low-level instructions for quantum control systems.

Orchestrated a multi-agent AI system with Amazon Bedrock AgentCore and FastAPI to automate Suspicious Activity Report (SAR) generation and provide counterfactual explainability for transaction anomalies.
Implemented real-time WebSocket streaming to feed risk metrics and multi-agent interactions into an interactive Three.js 3D spatial visualization frontend.

Redesigned the backend for a campus-wide dating event using MongoDB and Fastify.
Designed a weighted matching algorithm to pair users based on preferences and compatibility scores in real-time.
Handled secure user data management and authentication flows.

Authored a high-performance Rust client library for the LocalSend API, used by other open-source developers.
Developed a daemon and CLI, building deep understanding of Linux system services and process management.

Built a high-speed note-taking app with a Rust backend that parses Markdown to GraphQL in milliseconds.
Frontend built with Next.js featuring graph visualization and split-view editing.

Built ANGELA, an innovative Fintech agentic AI solution to detect money laundering, outperforming over 60 other competing teams.

Ranked 2nd of 150+ university teams; demonstrating expertise in attacking and defending multiple systems simultaneously.

Ranked 2nd out of 150+ secondary teams; displaying excellence in binary exploitation, reverse engineering, and cryptography.

Bachelors in Computer Engineering

International Baccalaureate