Events Pipeline activity log, diagnostics, and operational metrics.
Pipeline Events 11375 last 24h · 332069 last 7d
credit_exhausted: 169784 dlq: 98084 eval_success: 28637 eval_skip: 7479 rate_limit: 7383 rater_validation_warn: 6148 eval_failure: 4290 cron_run: 3967 eval_retry: 2301 model_divergence: 1822 dlq_replay: 791 self_throttle: 706 rater_validation_fail: 214 dlq_auto_replay: 187 coverage_crawl: 114 auto_retry: 41 trigger: 40 calibration: 28 cron_error: 28 rater_auto_disable: 15 story_dead: 4 story_flagged: 4 rater_auto_enable: 2
Recent Events
2026-03-02 10:56 eval_success Lite evaluated: Neutral (0.00) - -
2026-03-02 10:56 cron_run Cron: 0 new, 712 unique stories - -
2026-03-02 10:56 eval_success Lite evaluated: Neutral (0.00) - -
2026-03-02 10:56 eval_success Lite evaluated: Neutral (0.00) - -
2026-03-02 10:56 eval_success Lite evaluated: Neutral (0.00) - -
2026-03-02 10:55 eval_skip Anthropic credit pause: dispatching 100 stories to free models - -
2026-03-02 10:55 eval_success Lite evaluated: Neutral (0.00) - -
2026-03-02 10:54 eval_success Lite evaluated: Neutral (0.00) - -
2026-03-02 10:54 eval_success Lite evaluated: Neutral (0.00) - -
2026-03-02 10:54 eval_success Lite evaluated: Neutral (0.00) - -
2026-03-02 10:54 model_divergence Cross-model spread 0.34 exceeds threshold (2 models) - -
2026-03-02 10:54 eval_success Lite evaluated: Strong positive (0.64) - -
2026-03-02 10:54 eval_success Lite evaluated: Mild positive (0.10) - -
2026-03-02 10:54 eval_success Lite evaluated: Neutral (0.00) - -
2026-03-02 10:54 eval_success Lite evaluated: Moderate positive (0.56) - -
Recent Failures 5
Claude Sonnet 4.6 says it is 我是 DeepSeek when asked in Chinese (xcancel.com) claude-p failed (exit 1)
Human brain cells on a chip learned to play Doom in a week (www.newscientist.com) claude-p failed (exit 1)
I Just Cancelled My ChatGPT Pro Plan (aifordevelopers.substack.com) ingest failed: Validation failed: Expected 31 scores, got 10; Missing sections:
Samsung Galaxy update removes Android recovery menu tools, including sideloading (9to5google.com) claude-p failed (exit 1)
The use of hydroxyapatite toothpaste to prevent dental caries (pmc.ncbi.nlm.nih.gov) claude-p failed (exit 1)
Evaluation Queue 100 stories waiting
# Status Story Domain
1 pending Show HN: Commitdog – Git on steroids CLI (pure Go, ~3MB binary) aysdog.com
2 pending OpenBerth – Deploy AI-built apps and tools to your own server openberth.io
3 pending Show HN: Axiom – structural OCR for handwritten STEM notes www.useaxiomnotes.com
4 pending Show HN: Xpandas – running Pandas-style computation directly in pure C++ self
5 pending AWS ME-CENTRAL-1 Region Down (Due to additional loss of mec1-az3) self
6 pending Show HN: OxyJen – Java framework to orchestrate LLMs in a graph-style execution self
7 pending The Epstein Class Launches a War www.thebignewsletter.com
8 pending Show HN: Workz–Git worktrees with zero-config dep sync and a built-in MCP server github.com
9 pending Show HN: Steward – an ambient agent that handles low-risk work github.com
10 pending Show HN: I built an AI tool that walks you through Toyota's 5 Whys method www.fivewhys.ai
11 pending Show HN: Interactive 3D WebGL Globe for real-time daylight cycles github.com
12 pending Show HN: Visualize Git commit histories as animated force-directed graphs github.com
13 pending Show HN: I built open source Gmail organizer because I refused to pay $30/month github.com
14 pending Show HN: OpenTamago – P2P GenAI Tamagotch open.tamago.chat
15 pending Show HN: HN Bot Detector - Detects LLM-Generated Comments on Hacker News github.com
16 pending Show HN: AI agent that works autonomously while I'm offline hire-your-ai-guide.vercel.app
17 pending Floor113.com – A Scarcity-Driven Dating System Built on Deterministic Access floor113.com
18 pending Ask HN: When do you expect ChatGPT moment in robotics? self
19 pending How I Caught an Illegal Russian Spy www.youtube.com
20 pending Show HN: Sun – Realtime voice agent for group conservation not just turn taking www.getsun.io
21 pending Show HN: ClawShield – Open-source security proxy for AI agents (Go, eBPF) github.com
22 pending Show HN: LightJJ – Web-Based UI for Jujutsu VCS github.com
23 pending Show HN: MCP-firewall: I created a policy engine for CLI Agents github.com
24 pending Show HN: Shannon's Revenge – detect Claude in your codebase for DoD compliance github.com
25 pending Ask HN: How will most Anthropic customers respond to the supply chain risk? self
26 pending Researchers Deanonymize Reddit and Hacker News Users at Scale threatroad.substack.com
27 pending Show HN: Gala – Sealed types, pattern matching, and monads for Go github.com
28 pending Show HN: I built speedmux, a libghostty-powered terminal multiplexer github.com
29 pending Why are Chinese EVs cheaper than Tesla restofworld.org
30 pending Show HN: Service Book – Smart maintenance tracking for every car owner yourservicebook.com
31 pending IPv6-Mostly Networks: Deployment and Operations Considerations datatracker.ietf.org
32 pending Given AI, should I still consider becoming a computer programmer? Yes, and... htmx.org
33 pending Show HN: Ductwork – A Go platform for running AI agents on autopilot github.com
34 pending Show HN: I'm 15. I mass published 134K lines to hold AI agents accountable github.com
35 pending AWS UAE suffers AZ outage after "objects strike data center", amid Iran attacks www.datacenterdynamics.com
36 pending Psychology: Who dont maintain many close friends, learned independence too early www.bolde.com
37 pending Russia's war erasing Kostiantynivka's Soviet era mosaics kyivindependent.com
38 pending Kalshi voids some bets on Khamenei's ouster www.theverge.com
39 pending Khamenei Trade www.businessinsider.com
40 pending Brain, Think on Thyself knowablemagazine.org
41 pending Scientists train human brain cells on a microchip to play Doom www.dexerto.com
42 pending Show HN: Lightweight, S3-compatible object storage server with built-in web dash github.com
43 pending Ask HN: How to approach new people in 2026? self
44 pending Show HN: Updater – one command for macOS app updates github.com
45 pending Show HN: Deploybase – Compare GPU and LLM pricing across all major providers deploybase.ai
46 pending Show HN: Mrkd – A native macOS Markdown viewer with iTerm2/VSCode theme import github.com
47 pending Don't blame System76 because California wants to ruin Linux [video] www.youtube.com
48 pending Epoll's Handle-Centric Design vs. Kqueue's Filter-Centric Design (2021) ariadne.space
49 pending Show HN: PraxisJS – signal-driven front end framework and AI experiment praxisjs.org
50 pending The Panopticon Is Here: How the US Government Built an AI Superweapon matt728243.substack.com
51 pending A Day in the Life of an Enshittificator [video] www.youtube.com
52 pending 'Employers are increasingly turning to degree and GPA' fortune.com
53 pending Aura-State: Formally Verified LLM State Machine Compiler self
54 pending World Chess Championship matches have not had an on-board checkmate since 1929 en.wikipedia.org
55 pending Datacentre developers face calls to disclose effect on UK's net emissions www.theguardian.com
56 pending US Military reportedly used Claude in Iran strikes despite Trump's ban www.theguardian.com
57 pending Show HN: Boucle – A self-dogfooding autonomous AI agent framework in Rus github.com
58 pending 'Enshittification' blamed for fewer NZers feeling positive about the internet www.rnz.co.nz
59 pending FBI investigating 'potential nexus to terrorism' in deadly mass shooting www.cnn.com
60 pending Show HN: Spekkio: Reverse-engineer specs from vibe-coded apps github.com
61 pending Show HN: I built a tool that turns any API into a CLI for agents instantcli.com
62 pending 6 Practices that turned AI from prototyper to workhorse (106 PRs in 14 days) self
63 pending Show HN: I made a iron dome game deployclaw.com
64 pending Show HN: SkillMesh (role-based tool routing for Claude/Codex) github.com
65 pending How I approach vibe coding projects to make it not suck self
66 pending Norweigan Government comes out swinging on enshittification www.youtube.com
67 pending The mission to stop the next global backdoor before it starts thenewstack.io
68 pending Don't go to the shoe shop to buy plates naomialderman.substack.com
69 pending Show HN: MCP Playground – free MCP test servers, inspector, and 10K+ server list mcpplaygroundonline.com
70 pending The Norwegian Consumer Council delved into enshittification and how to resist it [video] vimeo.com
71 pending Show HN: Tree, but for Token Usage github.com
72 pending HN Skins – Read HN in Style with 5 Skins: Cafe, London, Teletype, Terminal, Nox github.com
73 pending Show HN: I'm a teen from Kenya and I built a pretty fast package manager in Rust github.com
74 pending Dutch Tax Authority hands US software company control over VAT system www.techzine.eu
75 pending Is AI Hiding Its Full Power? With Geoffrey Hinton [video] www.youtube.com
76 pending Simple Made Inevitable: The Economics of Language Choice in the LLM Era felixbarbalet.com
77 pending Show HN: Auto-cleanup for Claude Code's orphan process memory leak github.com
78 pending Show HN: I built a zero-browser, pure-JS typesetting engine for bit-perfect PDFs github.com
79 pending I used 2D Base64 to bypass Gemini and expose Google's moderation flaws self
80 pending Show HN: Rulegen – Auto-generate Claude.md and .cursorrules from your codebase github.com
81 pending Show HN: Vote-MCP" – a bit like Google Forms, but for AIs vote-demo.dapp32.com
82 pending A Day in the Life of an Enshittificator [video] www.youtube.com
83 pending Show HN: Terminal-Style Portfolio on the Internet kuber.studio
84 pending Show HN: Effective Git github.com
85 pending Intelligence is a commodity. Context is the real AI Moat adlrocha.substack.com
86 pending Show HN: Optimal: Cost effective infra with agentic inbox [video] github.com
87 pending Show HN: Built a tool that turns your GitHub commits into build-in-public posts www.smashlanding.xyz
88 pending He built a bar duty scheduling app for his Hockey club in a weekend medium.com
89 pending Latent-Space Communication in Heterogeneous Multi-Agent Systems arxiv.org
90 pending 390TB video game archive being taken offline due to skyrocketing RAM, SSD www.tomshardware.com
91 pending Ask HN: How do we solve the bot flooding problem without destroying anonymity? self
92 pending The Atlantic AI Watchdog www.theatlantic.com
93 pending Malm Whale in Gothenburg www.atlasobscura.com
94 pending Plugtest en.wikipedia.org
95 pending Banning children from VPNs and social media will erode adults' privacy www.newscientist.com
96 pending Why Apple's move to video could endanger podcasting's greatest power www.anildash.com
97 pending Tell HN: My daily game won a Players Choice Award self
98 pending Show HN: AIQuotaBar – See Claude/ChatGPT usage limits in your macOS menu bar github.com
99 pending NASA spots new signs of lightning on Mars www.scientificamerican.com
100 pending Ow My Foot – Error Handling Across C, Go, Rust, and Google's Absl n8z.dev
Dead Letter Queue 93 pending
Status
93 pending · 978 replayed · 97018 discarded
43691142 Whistleblower details how DOGE may have ...
43691142 Whistleblower details how DOGE may have ...
47184288 Leaving Google has actively improved my ...
47181178 JavaScript DRMs Are Stupid and Useless...
47137199 Modeling Cycles of Grift with Evolutiona...
API & Rate Limits 274932 errors 7d
Errors/warnings (7d)
API Headroom claude-haiku-4-5-20251001 · 2026-03-01 00:54
Requests
999/1000
Input Tokens
446k
Output Tokens
89k
Cache: 45% 429s: 0
Cycle Performance
Cycle Start Crawl Found New Evals Fail Cycle Time
2026-03-02 10:56 92.3s 712 0 2 --- current
2026-03-02 10:54 56.6s 712 2 14 --- 2m 35s
2026-03-02 10:51 62.5s 712 2 17 1 2m 57s
2026-03-02 10:49 56.4s 712 0 15 --- 2m 7s
2026-03-02 10:46 62.2s 712 1 16 --- 2m 53s
2026-03-02 10:44 57.6s 713 1 14 --- 1m 58s
2026-03-02 10:41 58.5s 712 0 17 --- 3m 6s
2026-03-02 10:39 57.4s 713 3 15 --- 2m 2s
2026-03-02 10:36 59.3s 714 2 14 --- 2m 57s
2026-03-02 10:34 58.6s 714 1 17 --- 1m 58s
Avg: 15.4 evals/cycle · 2m 30s cycle time
Throttle Impact 35m wasted 7d
Model Events Wasted (7d)
706 35m 22s
Eval Latency
Model P50 P95 P99 N
llama-4-scout-wai 7128.6h 96046.7h 127450.4h 2752
llama-3.3-70b-wai 7145.3h 96046.6h 127450.5h 2736
claude-haiku-4-5-20251001 91.1h 68278.4h 91146.6h 585