Amazon Web Services
Software Engineer Intern, Aurora Storage Performance
Designed adaptive runtime compression logic in Aurora's C++ read-replica path and extended Rust benchmarking tooling for read-replica clusters.
Nash Wu
Interested in AI infrastructure, CUDA, distributed systems, storage engines, and performance.
B.S. in Electrical and Computer Engineering at University of Texas at Austin, August 2023 - May 2027. Experience building storage-performance software at AWS Aurora and avionics-test systems at SpaceX.
Experience
SpaceX
Software Engineer Intern, Falcon and Dragon Avionics Test
Built avionics test systems software for telemetry, including a TCP/IP compute service, nonblocking event loop, fault-tolerant telemetry upload path, and ISO-SPI spoofer.
Solar Turbines
Software Engineer Intern, Internal Tools
Developed .NET engineering software for internal reporting, SQL-backed workflows, and automated test coverage.
National Taiwan University
Software Engineer Intern, Fiber Laser Communication and Nanophotonics Lab
Built software for reinforcement-learning-assisted PID tuning and peak finding in a fiber-laser control system.
Featured Projects
CUDA ML Kernel Runtime
A lightweight C++/CUDA inference runtime with graph compilation, reusable execution plans, custom transformer kernels, CUDA Graph replay, and runtime kernel selection.
Rust Distributed Key-Value Store
A distributed key-value store in Rust using Tokio, TCP RPC, Raft-style leader election, replicated logs, quorum commits, and local write-ahead logging.
Hybrid CPU-GPU Ray Tracer
A Whitted-style ray tracer implemented across OpenMP and CUDA, with custom BVH construction and GPU traversal optimizations.
Additional Projects
AWS Elastic Scale Orchestrator
A cloud training orchestrator that coordinates SageMaker jobs through a Spring Boot control plane, SQS workers, DynamoDB state, and AWS autoscaling services.
Real-Time Collaborative Document Editor
A cloud-native collaborative editor built with Go WebSockets, Angular, Yjs, PostgreSQL, Redis, and Kubernetes-based AWS infrastructure.