frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Token Spend Is Not a Productivity Metric

https://simdecisions.com/blog/token-burn-trap
1•daaaave-atx•1h ago

Comments

daaaave-atx•1h ago
Author here.

I've spent 30 years in operations management … call centers, workforce management, content moderation at scale (20K agents, 200 countries). Now I'm building AI orchestration tooling.

I almost fell for the token-burn metric myself. I'm between roles, and I caught myself thinking: if I can demonstrate high token spend, maybe that's my ticket to a big-salary AI engineering role.

Then I realized I was about to abandon everything I'd learned about measuring outcome instead of effort.

The Three Currencies framework (CLOCK, COIN, CARBON) and IRE fidelity come from that operations background. Happy to discuss, push back welcome.

Framework NextGen Event Announcement

https://frame.work/nextgen
2•starkparker•4m ago•0 comments

Datadog: We built a real-world evaluation platform for SRE agents at scale

https://www.datadoghq.com/blog/engineering/bits-ai-eval-platform/
2•SpaceJudas•5m ago•0 comments

Vercel Claude Code plugin wants to read your prompt

https://akshaychugh.xyz/writings/png/vercel-plugin-telemetry
3•akshay2603•7m ago•0 comments

Show HN: Cognitox – open-source Amazon Cognito emulator written in Rust

https://github.com/unvalley/cognitox
2•unvalley•9m ago•1 comments

Greece to ban social media for under-15s from next year

https://www.bbc.com/news/articles/ckgx1x742x5o
2•thunderbong•9m ago•0 comments

Claude mythos is the first model Anthropic didn't release

https://blog.alcazarsec.com/tech/posts/claude-mythos-the-first-model-anthropic-didnt-really-release
1•alcazar•10m ago•0 comments

The Dark Factory Harness: From Autonomous Hill-Climbing to Autonomous Research

https://sotaverified.org/blog/improving-autoresearch-dark-factory-harness
1•uberdavid•10m ago•1 comments

iPad at 16 – Redundant or Post-PC?

https://asymco.com/2026/04/09/ipad-at-16/
2•ndr42•10m ago•0 comments

How HTTPS Works

https://howhttps.works/
1•sebg•11m ago•0 comments

Workingasync.io – A job board for asynchronous remote jobs

https://workingasync.io
1•Log007•11m ago•1 comments

An Agent Skill that implements Karpathy's LLM-wiki on personal GitHub Repo

https://github.com/rarce/git-wiki
2•rarce•11m ago•0 comments

NASA Built Artemis II's Fault-Tolerant Computer

https://cacm.acm.org/news/how-nasa-built-artemis-iis-fault-tolerant-computer/
2•speckx•12m ago•0 comments

LLM agents shouldn't execute blindly – this one plans first and stays editable

https://cuddlytoddly.com/
2•philiparxist•12m ago•0 comments

Masjesu botnet targets IoT devices while evading high-profile networks

https://securityaffairs.com/190548/malware/masjesu-botnet-targets-iot-devices-while-evading-high-...
1•lschueller•12m ago•0 comments

Sociotechnical Archaeology

https://jensrantil.github.io/posts/sociotechnical-archaeology/
1•JensRantil•14m ago•0 comments

The M×N problem of tool calling and open-source models

https://www.thetypicalset.com/blog/grammar-parser-maintenance-contract
1•remilouf•17m ago•0 comments

Show HN: AI agents are bad at API integrations – we fixed it

https://www.apimatic.io/product/context-plugins/showcase
3•sohaibtariq•17m ago•2 comments

A Server That Forgets: Exploring Stateless Relays

https://blog.torproject.org/exploring-stateless-relays/
1•ahlCVA•18m ago•0 comments

Study: Google's AI Overviews show wrong answers every hour

https://www.popsci.com/technology/ai-overview-inaccuracy-google/
1•Brajeshwar•18m ago•0 comments

'Snoopy', 'Adolf' and 'Password': The Hungarian Government Passwords Exposed

https://www.bellingcat.com/news/2026/04/09/the-hungarian-government-passwords-exposed-online/
4•lschueller•19m ago•2 comments

Pipeline Architecture System Design

https://dvcoolarun.com/2026/04/09/Pipeline-architecture-system-design.html
2•dvcoolarun•19m ago•0 comments

Need Advice. Should I Give Up My CS Degree in the Last Semester?

1•jornbess•19m ago•0 comments

AI Code Security Scanner – static analysis for 14 languages, OWASP Top

https://github.com/astro717/ai-code-security-scanner
1•astro717•19m ago•0 comments

Using AI in Blogging

https://cagrimmett.com/2026/02/16/on-using-ai-in-blogging/
1•speckx•20m ago•0 comments

Real Time Tide Data Novaa

https://tidesandcurrents.noaa.gov/map/
1•prakashqwerty•21m ago•0 comments

A complete GPT language model in ~600 lines of C#, zero dependencies

https://github.com/milanm/AutoGrad-Engine
2•evo_9•21m ago•0 comments

What to Do with Your Extra Tokens

https://www.johnbhiggins.com/writing/extra-tokens
1•jhiggins777•21m ago•1 comments

Adam Back: I'm not Satoshi

https://twitter.com/adam3us/status/2041811857732768148
2•wslh•22m ago•0 comments

Uncharted island will soon appear on nautical charts

https://phys.org/news/2026-04-uncharted-island-nautical.html
1•pseudolus•23m ago•0 comments

Legos vs. 3D Printers

https://koaning.io/posts/legos-vs-3d-printers/
1•sebg•24m ago•0 comments