• Developing an LLM agent to generate efficient parallel versions of critical code, achieving a 30% improvement in matrix multiplication performance for torch.matmul• Implementing progressive optimizations including parallelization, vectorization, and other techniques for 40% greater speed• Utilizing CUDA, SLURM, and OpenMP for testing on NVIDIA GPUs, benchmarking optimized code against baseline implementations with a 20% performance gain

• Building 5+ AI apps and APIs using Next.js, OpenAI, Pinecone, StripeAI with 97% accuracy as seen by 1000 users• Developing projects from design to deployment leading 4+ engineering fellows using MVC design patterns• Coached by Amazon, Bloomberg, and Capital One engineers on Agile, CI/CD, Git, and microservice patterns

• Managing 5 teams of 5-7 Software Engineering interns developing specific sections of HokieHub• Integrating 2-week team sprints to address bugs, concerns, and readjusting targets• Developed a comprehensive plan to complete by late August 2024

• Measured neural network model efficiency utilizing ONNX runtime-optimized model for 10% greater scalability.• Designed increased scalability of neural network models for usage in ARM and RISC-V processors.• Optimized MySQL database queries, reducing page load times by 15% and enhancing overall application performance.

• Managed classes of 10 students teaching Python using dictionaries and function calling to create small-scale applications.• Optimized traffic for seamless parking resulting in $500 in violations saved weekly.• Maintained cleaning systems incorporating effectively dividing work between coworkers for 35% reduction in time to complete.