Jifeng Wu's Personal Website

Hugging Face Model Repositories: Organization, Semantics, and Portability

A practical mental model for what a Hugging Face model repository actually contains, where the model definition lives, and how to think about portability to separate runtimes.

2026-05-04

AI and Machine Learning

#reference #machine-learning #huggingface #model-deployment

Exporting Compute Graphs, LLM Shape Dynamics, and Serving Runtimes

Why CNNs export to one graph while LLMs need many, how Hugging Face hides this complexity through eager execution, and why vLLM-style systems act as specialized inference operating systems rather than

2026-05-03

AI and Machine Learning

#reference #systems #machine-learning #llm #inference #pytorch #vllm

Schedules in Machine Learning Computation: What They Are and Who Needs to Know About Them

A schedule is the plan for how a computation is executed efficiently on hardware. This post maps out the schedule-specification spectrum across ML frameworks, explains who needs to care about scheduli

2026-05-03

AI and Machine Learning

#reference #systems #machine-learning #computer-architecture

Important Locations on Jailbroken iOS

Important Locations on Jailbroken iOS

2026-04-26

Systems

#reference #systems #c++

Learning MLIR and HLO by Building a Tiny StableHLO-to-LLVM IR Compiler

This project explored how to lower machine-learning-oriented IR into low-level compiler IR by building a small MLIR-based compiler. The tool reads StableHLO MLIR, applies a sequence of lowering and cl

2026-04-26

Systems

#reference #systems #c++ #build-systems #machine-learning

2026-04 Financial Statement

2026-04 Financial Statement

2026-04-23

Statements

#records #finance

Using MLIR as a C++ Library with a Relocatable Install

Using MLIR as a C++ Library with a Relocatable Install

2026-04-19

Systems

#reference #systems #c++ #build-systems

Docker and Podman Containers as Lightweight VMs for Interactive Work

Docker and Podman Containers as Lightweight VMs for Interactive Work

2026-04-18

Systems

#reference #systems #linux #docker #podman

Jie Liu's B-Exam: Abstractions and Optimizations for Sparse Tensor Computation on Modern Hardware

A summary of Jie Liu's B-exam on sparse tensor computation, covering sparse format conversion, asynchronous GPU execution, and the design of high-performance BCSR/WCSR SpMM kernels.

2026-04-17

Research Notes

#machine-learning #llm #defense-summary #computer-architecture

Critique of "RFSeek and Ye Shall Find: A tool for summary visualization and analysis of RFCs"

A critique of RFSeek, arguing that provenance-linked RFC visualization is valuable but that the underlying methodology remains closer to a retrieval-augmented prompting pipeline than to an agentic, ve

2026-04-16

Research Notes

#systems #paper-reading #llm #visualization #software-engineering