Scribd
Upload
15 views1 page

Xida Ren: Education

Xida Ren is an expert in computer architecture, systems, security, and reliability with a PhD in Computer Architecture from the University of Virginia. They have extensive experience optimizing performance and lowering latency for distributed systems through hands-on work with languages like C++, Python, C, Zig, and frameworks like TensorFlow and PyTorch. Notable projects include discovering critical vulnerabilities in x86 processors, developing a memory subsystem for a parallel discrete event simulator, and achieving rapid cold starts for large language models.

Uploaded by

infraredhammerhead
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
15 views1 page

Xida Ren: Education

Xida Ren is an expert in computer architecture, systems, security, and reliability with a PhD in Computer Architecture from the University of Virginia. They have extensive experience optimizing performance and lowering latency for distributed systems through hands-on work with languages like C++, Python, C, Zig, and frameworks like TensorFlow and PyTorch. Notable projects include discovering critical vulnerabilities in x86 processors, developing a memory subsystem for a parallel discrete event simulator, and achieving rapid cold starts for large language models.

Uploaded by

infraredhammerhead
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Xida Ren

Expert in x86, RISC, Systems, Security, and Reliability. Fullstack / Cloud / Machine Learning Hobbiest

Education
2019 - 2023 PhD in Computer Architecture; University of Virginia
GRE: Verbal: 167, Quant: 168 / GPA: 4.0 / All but Thesis
2016-2019 BSc, Computer Science and Mathematics; College of William and Mary 3.8

Skills
Tools: Expertise in: C++, Python, C, Zig, SQL, MatLab, Haskell Familiar with: Pandas, Numpy, TensorFlow,
PyTorch, sklearn, Linux
Specialities: CPU/GPU Performance, Machine Learning, Queuing Theory, Risk, Statistics
Interests: Security, Hardware Accelerators, Formal Verification, Hearthstone, FengShui

Experience
2023: MevLink (Informal Consulting for Low-Latency Etherium Data Company)
• Cut version interval by half by consulting with team to implement test-driven development and agile practices
• Cut RLPx cryptographic overhead by 10x using LLVM and hand-coded AVX512 Keccak-256 implementation.
• Lowered data latency (company’s key competitive advantage) for distributed Eth data system by 2x in my first week by
mastering Ziglang from scratch to optimizing a 70,000 LoC codebase.
Aug 2022-Jan 2023: Intel Labs, Research Intern
• Achieved <3% CPI and <10% MPKI error while speeding up benchmarking by 10^6x using SimPoints, measured over
SPEC 2017 x86 workloads
• Sped up new-workload SimPoint generation 200x by sampling CPU performance counters to generate SimPoints,
avoiding simulation and instrumentation overheads.
• Use differential privacy to enable trace-sharing across organizational boundaries without concern for leaking sensitive
IP.
May 2022 - Aug 2022: NXP Semiconductors, Hardware Security ML Engineer
• Achieved 85% accuracy rate in generalizing detectors to zero-day attacks on edge hardware by training unsupervised
ML algorithms.
• Compared supervised and unsupervised algorithms for Spectre Side-Channe Detection, including VAEs, logistic regres-
sion, perceptrons, time-convolutional neural networks, decision trees, k-nearest neighbors, random forests, and support
vector machines.
• Designed system in close collaboration with the VP of Edge Software for usable tradeoffs between detection accuracy
and performance.
Aug 2019-May 2023: UVA Computer Science Department, PhD Candidate (Exited All-But-Thesis)
• Unearthed two critical vulnerabilities in modern x86 processors, enhancing security and integrity, published in ISCA
2021
• Wrote LLVM / MLIR compiler passes for discovering vulnerabilities and optimizing / studying Machine Learning Models.
• Applied formal verification techniques to harden machine learning models against adversarial attacks.
Aug 2020 – Nov 2020: Lawrence Berkeley National Lab, Research Intern
• Contributed to the PARADISE++ Project, spearheading the development of a memory subsystem for a parallel discrete-
event simulator with C++ and Valgrind.
• Engineered an optimistically synchronized parallel discrete-event simulator memory subsystem, elevating project effi-
ciency and output.

Selected Projects
2023 LLM Quick Serverless Inference Project – Achieving Rapid Cold-Starts with Large Language Models
• Implement REST-based Upload-Your-Own-Model serverless inference infrastructure using FastAPI
and Docker
• Achived cold-start latency 4x faster than industry standard (15s, down from 67s-3m59s for GPT Neo
1.3b) using AWS SageMaker (compute), AWS S3 (model caching), Docker-Compose (orchestra-
tion), and Ansible (deployment)
2017 Int’l Genetically Engineered Machines Contest (iGEM 2017) Won Best Model award by modeling genetic
circuit behaviors and designing novel gene expression rate control methods.
2017 React-HexLife – Conway’s Game of Life on Hexagonal Grid, Available on Github - Developed a
distinctive version of Conway’s Game of Life using a hexagonal grid, innovating on a classic concept
using React.js - Implement robust full-stack solution with NodeJS for backend development - Revisiting
and enhancing the project with “bun” framework

Contact
xida.ren.career@gmail.com • (757) 279-4582 • Website • Philadelphia, PA

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy