LLMs on a Shoestring: The Dynamic Cache Advantage by Arvind Sundararajan
dev.to·1d·
Discuss: DEV
💾Cache Algorithms
Cost-Effective, Orthogonal Approach to Resilient Memory Design (Univ. of Central Florida, UT San Antonio, Rochester)
semiengineering.com·8h
🧠Memory Models
Crashes are loud. Leaks are quiet.
blog.bitdrift.io·1d
📊Memory Profilers
Semantic Dictionary Encoding
falvotech.com·1d·
Discuss: Hacker News
🗂️Type Indexing
XMP/EXPO profiles aren’t just for faster speeds; here are 3 other advantages I get
xda-developers.com·1h
Performance
Safepoints and Fil-C
fil-c.org·10h·
Discuss: Hacker News
🎯Ring Buffers
The Unseen Variable: Why Your LLM Gives Different Answers (and How We Can Fix It)
hackernoon.com·6h
🗺️Region Inference
How to implement the Outbox pattern in Go and Postgres
medium.com·32m·
Discuss: Lobsters
📮Message Queues
Machine Scheduler in LLVM
myhsu.xyz·7h·
Discuss: Hacker News
📅Instruction Scheduling
Erase and Rewind: Precise LLM Memory Manipulation for Safer AI by Arvind Sundararajan
dev.to·7h·
Discuss: DEV
🔗Memory Linearization
Building a Simple Stack-Based Virtual Machine in Go
blog.phakorn.com·1d·
📚Stack Data Structures
Arguzz: Testing zkVMs for Soundness and Completeness Bugs
arxiv.org·11h
🧪Testing Compilers
Rowhammer: TRR on DDR5 DRAM has been broken
comsec.ethz.ch·22h·
🏷️Memory Tagging
Predictive Precision: Combining Data and Reasoning for Self-Healing Systems by Arvind Sundararajan
dev.to·2h·
Discuss: DEV
🚂Error Propagation
Ask HN: How can I test FTS5 engine in SQLite3?
news.ycombinator.com·6h·
Discuss: Hacker News
Performance
What I Learned During My Paternity Leave
thecoder.cafe·5h
📚Self-Documenting Code
AQUA: Attention via QUery mAgnitudes for Memory and Compute Efficient Inference in LLMs
arxiv.org·11h
🔍ML Language
Building Production-Ready Raft Consensus in Golang: Implementation Lessons and Performance Optimizations
dev.to·5h·
Discuss: DEV
📡Erlang BEAM
Rust-style safety model for C++ 'rejected' as profiles take priority
theregister.com·2h·
Discuss: Hacker News
🌈Gradual Languages