sumi.news
Search
Following
Sign in
← Back to news
Programming
RSS Feed
sumi.news
Programming
Follow
Latest
Thu May 7
How I built local-first memory for Claude Code, Cursor, and Codex - 94.5% LoCoMo recall@10, 70ms p50
6d
H
AI Is Not a Football Team: Why I'm Still Trying to Figure This Out
6d
H
Why Institutional Credit Is Moving to Trade Finance Rails
6d
H
Four Reasons Why Strategic QA Is a Fintech Superpower
6d
H
Programmable payment Recovery in Failing Networks
6d
H
The Great American Grid - Rebooted
6d
H
Sysadmining like it's 2009
6d
L
How Binance Is Closing the $2B Infrastructure Gap Between TradFi and Crypto: Institutional Reporting
6d
H
The Engineering Leader’s Guide to AI Tools for Software Development
6d
H
The Greece Directive
6d
H
The Hidden Insanity of DynamoDB Pricing
6d
H
Why We Built LiveView: Moving FastNetMon from the Terminal to the Operational Surface
6d
H
Windy Colony
6d
H
Singapore’s Agnes AI Becomes First Singapore AI Lab on Global Benchmark Leaderboard
6d
H
Five Years of Trying to Add Recursion to lychee
6d
L
zsh 5.9.1 released
6d
L
How to Build a Shared Lint Package for Every Repo
6d
H
The Real Moat Isn't Software
6d
H
Bad Specs Are Bureaucracy. Good Specs Are AI Leverage
6d
H
The Next Bottleneck in AI-Assisted Engineering Isn’t Code
6d
H
The AI Agent Stack I’d Standardize Around Today
6d
H
Five Ways to Lower Inference Costs in AI Products
6d
H
UTF8 email with DMA: DragonFly Mail Agent
6d
L
It's Not Just X. It's Y
6d
L
Kernel prepatch 7.1-rc6
6d
W
On Reading SRAMs in IR Images, and Establishing Bounds on Trust
6d
L
The software industry: annealing, but wrong
6d
L
Concurrent device registration without Redis
6d
L
Why does ASTC use ISE when almost nothing else does?
6d
L
Homomorphic static analysis
6d
L
Five Common Failure Modes in FLUX Illustration LoRA Training
6d
H
Are New Turing Tests Measuring Intelligence or Human Anxiety?
6d
H
New AI Benchmarks Are Testing Consistency Instead of Memorization
6d
H
llms-txt-gen Earns a 49.78 Proof of Usefulness Score by Building an AI-Readiness CLI Tool
6d
H
Claude Code Works Better When You Let Sessions Die
6d
H
What Two Years of Research Have Taught Us About Chunking for RAG
6d
H
Adding Offline Mode and Custom Servers to an MMORPG
6d
L
275 Blog Posts To Learn About Growth
1w
H
What AI Can't Measure About Human Potential
1w
H
vibecoding as adhd multiplier
1w
L
Every Team Makes These Four Decisions When Using Git
1w
H
Unlawful by design: Exposing the human rights costs of generative AI
1w
L
Blorp Language
1w
L
The HackerNoon Newsletter: The Death in the Middle (5/31/2026)
1w
H
Streaming ZIP Archives On the Fly With nginx + mod_zip: No Disk, No Buffers, No Problem
1w
H
Front-End’s Missing Metric: The TBT Window
1w
L
SOPS + Age and Sealed Secrets
1w
L
Croft - TUI VSCode Clone
1w
L
The TechBeat: AGENTS.md Was Built to Help Agents. I Use it to Catch Them. (5/31/2026)
1w
H
Navigating Claude Code: Skills That Actually Fire
1w
H
More →