My AI Workflow for Understanding Any Codebase
steipete.me·19h
⚡Proof Automation
AST, Bytecode and the In Between: An Exploration of Interpreter Design Tradeoffs
2025.ecoop.org·5h·
Discuss: Hacker News
🔤DSLs
Deep Dive into Databend UDF, implementing your data solutions with Python, WASM
databend.com·15h·
Discuss: Hacker News
📋DFDL
Rust: A language that grows with you, your career and your projects
kerkour.com·1d·
Discuss: Hacker News
🦀Rust Macros
Getting an LLM to set its own temperature
amanvir.com·1d·
Discuss: Hacker News
🐛Fuzzing
Memory Safety in Web Rust System Zero Cost Secure（1750594760521200）
dev.to·3d·
Discuss: DEV
🦀Rust Borrowing
Perspectives in Play: A Multi-Perspective Approach for More Inclusive NLP Systems
arxiv.org·2h
📚Digital Humanities
Programming by Backprop: LLMs Acquire Reusable Algorithmic Abstractions During Code Training
arxiv.org·2d
💻Local LLMs
Flame Graph Performance Truth Analysis（1750839255230100）
dev.to·21h·
Discuss: DEV
⚡Performance Tools
Black-Box Test Code Fault Localization Driven by Large Language Models and Execution Estimation
arxiv.org·1d
🔍Concolic Testing
This Is How I Mastered TypeScript Like I'm 5 (And How You Can, Too!)
dev.to·2d·
Discuss: DEV
🔗Dependently Typed
LangChain vs. TLRAG: A Comparative Analysis for Investors
dev.to·1d·
Discuss: DEV
🌀Brotli Internals
Compbolt: A lib with a hard to misuse API (based on Matt Godbolt)
github.com·21h·
Discuss: Hacker News
🌀Brotli Internals
Fine-Tuning and Prompt Engineering of LLMs, for the Creation of Multi-Agent AI for Addressing Sustainable Protein Production Challenges
arxiv.org·2h
🧮Theorem Proving
A large deviation view of \emph{stationarized} fully lifted blirp interpolation
arxiv.org·1d
📐Compression Bounds
Modeling the World in 280 Characters
tympanus.net·2d·
Discuss: Lobsters, Hacker News
⟷Bidirectional Programming
Contextualizing SUTRA: Advancements in Multilingual & Efficient LLMs
hackernoon.com·14h
💻Local LLMs
Meet Mojo: The Language That Could Replace Python, C++, and CUDA
hackernoon.com·22h
⬆️Lambda Lifting
From Codicology to Code: A Comparative Study of Transformer and YOLO-based Detectors for Layout Analysis in Historical Documents
arxiv.org·2h
📃Manuscript Tokenization
SPD-CFL: Stepwise Parameter Dropout for Efficient Continual Federated Learning
arxiv.org·2d
📊Learned Metrics