Run 20261021T045117Z
Oct 21, 2026, 4:51 AM UTCResult
QUOTA
Duration
1m 26s
Tokens
6.9k
2 LLM calls · day total 24.9k (6% of 400k)
Valid startups
2
8 people
Quality gates failed
- only 2 valid startups (< 3)
- only 1 discovery sources produced valid startups (< 2)
Stage timeline
- discover8.1s
- triage20s
- crawl17s
- extract38s
- gates0.0s
Funnel
- found████████████████████████████306
- triaged█░░░░░░░░░░░░░░░░░░░░░░░░░░░93%
- crawled█░░░░░░░░░░░░░░░░░░░░░░░░░░░889%
- extracted░░░░░░░░░░░░░░░░░░░░░░░░░░░░225%
- relevant░░░░░░░░░░░░░░░░░░░░░░░░░░░░2100%
- valid░░░░░░░░░░░░░░░░░░░░░░░░░░░░2100%
Source health
- yc██░░░░░░░░░░░░░░░░12 · 0.55s
- hiring██████████████████120 · 0.96s
- hn██████████░░░░░░░░67 · 1.9s
- web█████░░░░░░░░░░░░░30 · 5.4s
- rss███████████░░░░░░░71 · 5.7s
- edgar█░░░░░░░░░░░░░░░░░6 · 8.07s
Planner output
web queries
- › seed round semantic search startup founded 2024
- › Series A RAG platform launch founder interview
- › AI retrieval tool pre‑seed funding Lightspeed Ventures
- › developer tooling startup vector search seed funding 2024
- › knowledge base AI startup Series A investor Andreessen Horowitz
hacker news keywords
semantic search · RAG · vector database · developer tooling · next.js AI
What changed
not recorded
Runs recorded before 2026-11-06 did not log a diff; and a run that did not pass stores nothing.
Quality
validity 100%
- avg score
- 73.5
- ungrounded people/links dropped
- 0
- pages gated
- 0
- with website / email
- 0 / 0
- people linked to a profile
- 6
- deferred (token budget)
- 0
issues: warn: empty tech stack ×1 · extraction error ×1
Crawl budget feedback
yield over the last runs → this run's caps
hn████░░░░░░░░░░░░3/11 valid · avg score 50 · cap 2
yc████████████████9/9 valid · avg score 54.9 · cap 2
rss█████████████░░░9/11 valid · avg score 74.9 · cap 2
web█████████████░░░9/11 valid · avg score 79.3 · cap 2
launch_hn████████████████4/4 valid · avg score 69.5 · cap 3
edgar███░░░░░░░░░░░░░1/6 valid · avg score 66 · cap 2
hiring██████░░░░░░░░░░2/5 valid · avg score 65 · cap 2
Decision engine
jev
153 decisions at 134 ms each · 0 pages skipped before extraction (~0 LLM tokens saved) · Jev input tokens: 72783
Accepted (2)
- Jenz via rssscore 72 · 3 people · 6 connections · 0 email · 0 phone
missing: tech stack, email, phone
- Plinto via rssscore 75 · 5 people · 5 connections · 0 email · 0 phone
missing: email, phone
Rejected (7)
- https://www.sec.gov/Archives/edgar/data/2144077/000214407726000001/[edgar] no company website found
- https://techcrunch.com/2026/02/25/gushwork-bets-on-ai-search-for-customer-leads-and-early-results-are-emerging/[web] extraction error
- https://finance.yahoo.com/sectors/technology/articles/exclusive-andreessen-horowitz-backs-deeptune-120000793.html?fr=sycsrp_catchall[web] extraction error
- https://news.ycombinator.com/item?id=49961408[hiring] extraction error
- https://joinpond.ai[hiring] extraction error
- https://www.ycombinator.com/launches/W5J-scout-by-fini-an-ai-teammate-for-your-support-team[yc] extraction error
- https://www.ycombinator.com/launches/W49-antropi-robotics-cnc-parts-in-days-from-autnomous-factories[yc] extraction error
Error log (1)
extract failed https://techcrunch.com/2026/02/25/gushwork-bets-on-ai-search-for-customer-leads-and-early-results-are-emerging/: Error code: 429 - {'error': {'message': 'Rate limit reached for model `openai/gpt-oss-120b` in organization `org_01m5bf9cd3epf9ytwbpr8r3nm7` service tier `on_demand` on tokens per day (TPD): Limit 200000, Used 196683, Requested 3729. Please try again in 2m57.984s. Need mo