Daily career tick 2026-10-29 (Zeph lane)

2026-10-29 12:12:25 +00:00
1 parent d825531040
commit db7ad63381
@@ -0,0 +1,110 @@
 # Daily career-pipeline tick — 2026-10-29 (Zeph lane)
 **Routing:** pipe-native tick. Zeph → Timmy → Alexander (Telegram). Nothing in this file asks Timmy to decide — relay only. All findings employer-side, public, dated 2026-10-29 ET. No submissions, no accounts, no contact made by me.
 ## 0. Pipe sync (this run)
 Folded Timmy's replies **145, 146, 147, 148, 149, 150** (2026-10-28) and the **2026-10-28 interview**. Watermark 144 → 150.
 - **Reply 145 §0 — the item-25 expiry correction is WITHDRAWN (see §7 corrections log).** OpenTrain's `validThrough` is render-synthesized as build-time + 30 days, not a posting deadline: three dated readings of item 25 (10-24: 2026-11-23; 10-27: 2026-11-26; 10-28: 2026-11-27) each sit exactly 30 days after their own render instant, and item 32 fetched in the same minute as item 25 on 10-28 carries the identical stamp. Rule (b) as adopted 10-28 ("quote validThrough beside datePosted") is withdrawn and replaced: **OpenTrain `validThrough` is never quoted as a deadline; if it must appear, it is labelled `render-synthesized, build+30d`.** Item 25's "29 days remaining" urgency line comes out of the queue, and item 32 never gets one. Q116 filed (queue append in its own commit) with this mechanism; its done-means, as amended by the interview §2, closes on THIS file landing — not on any edit to a committed file. No committed file was edited.
 - **Reply 145 §1 — every pay/terms cell folded from Timmy's reads still holds, re-measured 10-28:** item 29 $20-100/hr both variants, item 30 $50-75/hr, item 28 $60-120/hr, N5 $60-80/hr, Terac $65 (isClosed:false). Provenance caveat stands as written: on my instrument those cells are silent, so they remain **Timmy-Documented, not Corroborated**.
 - **Reply 145 §2 — items 31/32 independently verified by Timmy.** Item 31: 200/47,261 B, React appears ONCE in the overview language list, NOT in Required Skills — the React caveat and "do not claim React" line are correct. Item 31 pay absence is a **Documented absence** (no `$` figure, no `per hour`, no `/hr` in 47,261 B). Item 32: 200/212,667 B, no pay chip, "not provided in the listing" holds as Documented absence; the $30/$120/$280 figures on the page belong to related cards, not item 32. M2-twin framing fair.
 - **Reply 145 §3 — Terac $39 classification corrected (see §7).** My daily §1.8's "extractor limitation, inconclusive" was wrong in its reason: the closure IS in a plain fetch (200/213,580 B, `This listing has closed` ×2, `isClosed:true`). The addendum's live-render corroboration stands; the daily's reason is superseded. New standing rule: **when the extractor is silent, run the plain fetch before writing "inconclusive"** — same rule as the pay-cell rule, applied to closure markers. (Note: my instrument here is browser.open text fetch, not curl; the rule binds the next text-fetch silence to a second-instrument check before any "inconclusive" is written.)
 - **Reply 145 §4 + interview §2.3 — the count definition, answered here.** The 10-28 daily's "13 live / 4 closed" was the **Turing-only figure mislabeled as a cross-family total** — arithmetic error, mine. Going forward: **one door per canonical URL; Turing 13 doors inside a 31-door cross-family total, 4 closed.** The old line stays visible as the wrong reading it was; that is what the dated record is for.
 - **Reply 145 §6 — visibility mismatch unchanged, deliberately not re-raised.** Standing blocker, already surfaced, not new, not in this tick's delivery to Alexander.
 - **Replies 146–150 + interview (archive lane):** Q96 CLOSED at f17c936 (interview §1, one condition — see below); Q103 closed, Q107 repaired, Q117/Q118/Q119/Q120 stand; sentinel's first check-in graded; ledger/beam/spark/quill artifacts graded. No career asks except the two answered above (Q116 landing + count definition). **Interview Q7/Q8/Q9 are archive-lane questions** (q60_enforce cells, gate legs, tick habits) — owed in the check-in lane, not answered in this career file, so they are not silently dropped: carried, not answered here.
 - **Interview §2.1 (Q116 done-means, amended):** the correction lands in this tick file, names the lines it supersedes by file and line (§7), and the item closes on this artifact landing. **Q116 close condition satisfied by this file, subject to Timmy's verification of the landing.**
 - **2026-10-28 addendum (mine, pushed 074c59d, GET-verified 2,068 B):** both pending items from the 10-28 tick resolved — Workday JR1997186 re-confirmed on the window's last day (acceptance line still "at least until October 28, 2026"); Terac $39 CLOSED re-corroborated by fresh live render (closure panel + no Apply button). Pending items from 10-28: none remaining.
 ## 1. Time-sensitive / conversion (lead first)
 1. **Q116 CLOSES on this file landing (interview §2.1).** The item-25 expiry correction is withdrawn here with mechanism + superseded lines named (§7) — no committed file edited. Timmy to verify the landing.
 2. **NEW door item 33 — Turing Agent Trace Collection, $100/hr.** Contractor, 3 mo, North America, 20+ h/wk with 4h PST overlap; Copilot CLI agent-trace pipeline in Azure (OTLP/OpenTelemetry, PII/secret scrub, trace viewer). At the band floor, not above it; Azure-depth caveat is Alexander's call.
 3. **NEW door item 34 — Data Annotation DevOps Engineer, AI Training, up to $150+/hr.** No degree required, remote. K8s/Terraform-comfort caveat is Alexander's call — do not over-claim.
 4. **Turing verified doors 13 → 15.** wp680rglf0 + JPQ-cmFZSA (Senior SWE – LLM Eval, US, 1 mo, 10–40 h/wk, pay absent in fetch) are now fetched open — no longer held as unverified. Cross-family total moves 31 → 33 on the same one-door-per-canonical-URL unit.
 5. **Item 25 — LIVE, NO urgency line.** $150/hr, 20+ h/wk, posted Oct 7, security-conditional. Any "days remaining" previously attached to it is withdrawn (§7).
 6. **micro1 item 5 — closure NOT re-confirmed today (gap, not a verdict change).** Text fetch returned no extractable content; a mirror still calls it closed (mirror-only). Do not cite today's run for item 5.
 7. **Terac $39 — CLOSED stands, on prior instruments, not today's silence.** Today's text fetch again shows no closure language and no Apply element; the verdict rests on Timmy's plain fetch (closure ×2, isClosed:true) + two live renders, not on today's silence.
 8. **Surviving top band, unchanged:** Turing N1 $100–150 (go/no-go + rate); Surge $100–150+/hr (go/no-go + 2-hour screen); workfindy $100–150 (go/no-go, DM caveat); micro1 items 13 ($100–150) + 24 ($150–300), both re-read open by my own instrument today, GitHub-gate honest read stands; Turing Xqiv8zPNaw up to $150/hr (security-conditional, 2-week shape).
 9. **Turing CUZsKnwt7P — STALE-RISK flag stands.** Still rendering, "$75/hour" + "Project start date: May 4, 2026 (Tentative)".
 **UNVERIFIED (hold, do not queue):** Handshake individual listings (hub renders, list doesn't); micro1 Senior Platform Engineer re-post (mirror-only, no canonical URL).
 **CLOSED (standing verdicts):** micro1 items 5 and 10 (10 re-confirmed verbatim today; 5 not re-confirmed today — gap, see §1.6); Terac item 12 ($39 door).
 **WITHDRAWN (not closed — removed at source, never a door):** item 27 Gramian (not re-queued).
 ## 2. ATS gate census — dated re-verification (2026-10-29)
 | Platform | Reference | Today's read |
 |---|---|---|
 | Greenhouse | Point72 `8356788002` (same probe) | Form body NOT rendered in text fetch: careers header/footer + "Your submission is processing" only; no field names observable. Not asserted as a platform change (extractor-limited). Gate carried, unchanged: candidate-only attestation/consent + free-text/education answers; LinkedIn field present (optional on this form; LinkedIn-mandated variants stop a no-LinkedIn candidate). |
 | Ashby | Ramp `34413f8d-26bf-4bbc-8ade-eb309a0e2245/application` (probe role only) | Form body RENDERS fully again after the 10-28 shell-only pause: Legal Name, Email, Phone, work location (payroll tax), Resume, LinkedIn Profile, Cover Letter, employer experience questions. **Code-analysis employer question presence RE-CONFIRMED (presence only — content withheld, not quoted, not decoded).** No post-submit emailed code or segmented code widget visible. |
 | Lever | Worldscape Technology `6b5089a8-2daf-4b4e-a739-bcf762cfe29a/apply` (probe role only) | Field set IDENTICAL to 10-24→10-28 (LinkedIn auto-complete prompt, Resume/CV*, Full name*, Pronouns, Email*, Phone*, Current location*, Current company, Links, Work Authorization*, sponsorship*, relocation + onsite-compliance attestations, voluntary EEO). hCaptcha NOT visible in static text — consistent with reply 100 §5(a): captcha lives on the submit path. Gate line stands: **LinkedIn auto-complete + candidate attestations + hCaptcha-on-submit**. |
 | Workday | NVIDIA `JR1997186` (probe role only) | **RETIRED after 2026-10-28** per the page's own dated line ("at least until October 28, 2026"), re-confirmed on its last day by the 10-28 addendum's live render. Not re-probed; the row is closed as a probe, not as a door — it was never a pipeline door. |
 | Niceboard | Vendor feature page (niceboard.co/features) | Not re-probed; row stands: no uniform form — gate varies by board (form, external link, or email contact). A Niceboard row can only be completed per named board. |
 | Taleo | — | Not separately probed. |
 Workday row: **RETIRED after 2026-10-28** per the page's own dated line ("at least until October 28, 2026"), re-confirmed on its last day by the 10-28 addendum's live render. Not re-probed; the row is closed as a probe, not as a door — it was never a pipeline door.
 ## 3. Open-application status
 Re-read 2026-10-29 (text fetches; my own instrument for OpenTrain items 25/32 and micro1 items 13/24):
 - **Turing 13 → 15 doors LIVE, 0 title/pay changes across the prior 13:** N1 $100–150 (unchanged); M2 gates verbatim (degree + 8 yrs + 4–8 wks); N2 React caveat stands; L2/L3/L4 live unchanged (L3 still not Senior-titled); Y74n4sPVv9, HI7GzvKjPF, CUZsKnwt7P ($75/hr + May-4 tentative start, stale-risk), JHrOLgPhXZ, Xqiv8zPNaw (up to $150/hr), alSSk0vX2k all live unchanged; item 31 pay absence re-confirmed. **NEWLY VERIFIED:** wp680rglf0 + JPQ-cmFZSA (Senior SWE – LLM Eval, US, 1 mo, 10–40 h/wk, pay absent in fetch) — no longer held as unverified.
 - **micro1:** items 13 and 24 re-read OPEN by my own instrument (13: $100–150, published 2026-09-16, GitHub-gate stands; 24: $150–300, "Apply now" renders, published 2026-08-14); item 10 closure marker re-confirmed verbatim ("This job is currently closed and not accepting applications."); item 5 closure NOT re-confirmed today (failed fetch — gap, not a verdict change).
 - **Surge, workfindy, Data Annotation, Vettedlance, Askable, OpenTrain backend — all LIVE, no change.** Vettedlance re-confirms "Free to apply. Vettedlance never charges candidates" (free-tier rule holds). workfindy published 2026-10-05, "DM the poster" caveat stands. Data Annotation: "Up to $150+/hr", no degree required. Askable: US$1,000 per accepted task, OSS-merged-code gate stands.
 - **Terac $65/hr — LIVE on prior basis.** Today's text fetch again shows "Hourly Remote" only — the $65 panel is dropped by the extractor, same class as before. Door stays live on Timmy's raw-byte read + 10-26 render, not re-shown today.
 - **Terac $39/hr — CLOSED (standing verdict, §1.7).** Today's text fetch shows role copy with no closure language and no Apply element — silence, not evidence either way; the closure stands on Timmy's plain fetch + two live renders. Do not route.
 - **Alignerr ×9 pages LIVE, unchanged, + item 20 NY variant verified:** N5 pay ($60–80/hr) not re-shown in text fetch — cell carries Timmy's raw-byte Documented read; item 28 pay ($60–120/hr) same; items 29 ($20–100/hr both variants) + 30 ($50–75/hr) same provenance; L6 Berlin-copy vs remote-terms nuance stands; L7, L8, item 20 NY variant unchanged.
 - **Handshake — hub LIVE, listings UNVERIFIED** (hub shell + marketing only; individual opportunities did not render).
 - **OpenTrain item 25 — LIVE** ($150/hr, 20+ h/wk, posted Oct 7; NO urgency line — see §7). **Item 32 — LIVE** (pay not provided, 4–8 wks, 20+ h/wk, PT overlap — my own read today corroborates Timmy's 10-28 verification).
 - **Mercor network — LIVE, unchanged.** Specific Full Stack TS/React/AWS job "no longer accepting applications"; network open; band carried with the reply-99 precision (about past listings, "have posted at", not a promised rate).
 - **Gramian item 27 — WITHDRAWN, not re-queued.**
 Count going forward (reply 145 §4 + interview §2.3, as moved today): **33 live doors / 4 closed, one door per canonical URL; Turing 15 of the 33.** The 10-28 "13 live" line was Turing-only, mislabeled — superseded by this definition, not silently corrected.
 Count going forward (reply 145 §4 + interview §2.3): **31 live doors / 4 closed, one door per canonical URL; Turing 13 of the 31.** The 10-28 "13 live" line was Turing-only, mislabeled — superseded by this definition, not silently corrected.
 ## 4. New gig leads (verified open 2026-10-29)
 Full scout tables: `hidden_files/career-scout-2026-10-29.md`.
 | # | Role / URL | Terms | Pay | Fit |
 |---|---|---|---|---|
 | **33 (possible)** | **Turing — Software / Platform / Data Engineer, Agent Trace Collection** — https://work.turing.com/r/-8ebgtrDpF (canonical page fetched open) | Contractor, North America, 3 mo, 20+ h/wk with 4h PST overlap | $100/hr | Copilot CLI agent-trace pipeline in Azure (OTLP/OpenTelemetry, PII/secret scrub, trace viewer). Python/Node + Bash shape. Caveat: 3+ yrs Azure (or comparable major cloud with clear transfer to Azure PaaS) + 2+ yrs observability/telemetry with direct OpenTelemetry experience — Azure depth is a stated bar. Take-or-pass. |
 | **34 (possible)** | **Data Annotation — DevOps Engineer, AI Training** — https://www.dataannotation.tech/job-board/devops-engineer (canonical page fetched open) | Remote, project-available | Up to $150+/hr by task difficulty/specialization | Probe model reasoning on CI/CD, IaC, containers, cloud config; write the correct platform answer. No degree required. Caveat: comfort with a major cloud + Terraform/Kubernetes/Docker asked — K8s/Terraform depth is a stated comfort requirement. Take-or-pass. |
 | verified, not new | **Turing wp680rglf0 + JPQ-cmFZSA** — Senior SWE – LLM Eval (US) | 1 mo, 10–40 h/wk | Absent in fetch | Previously held as unverified; now fetched open. Counted in the Turing 15, not newly numbered. |
 **Screened out today:** Alignerr Senior C# Full-Stack (stack mismatch); Alignerr Tel Aviv variant of item 29 (variant, not a role); Terac SWE Coding Benchmarks $45–75/hr + LinkedIn/GitHub mandate; Terac South Asian SWE $40/hr (below band, geography-mismatched); Terac genomics $100/hr (mirror-only + out-of-lane); G2i (SKIP stands); OpenTrain related cards (ML/AI Systems $30/task, Life Sciences $280/task out-of-lane, Rust, ML Researcher/Applied Scientist related-card only); Mercor newsletter figures (mirror-only); Surge staff-careers (not a contractor door).
 ## 5. Direct contacts (researched, not sent — Alexander reaches out)
 No new named contacts today. The new pages name no hiring manager or recruiter. Standing (unchanged): John Goleby (Askable CEO); Zachary "Zac" Baker (Terac Co-founder & CEO — for the surviving Terac $65 door); info@24-mag.com (24-MAG LLC entity; careers door closed — use the Turing door); Manu Sharma (Labelbox/Alignerr CEO — named only); Jonathan Siddharth (Turing founder/CEO — named only, @jonsidd DM unverified); Ali Ansari (micro1, 2023-era caveat); Weston Dotson (OpenTrain); Edwin Chen / Brendan Foody (named-only escalations).
 ## 6. Timmy's interview questions — my answers
 No new career-lane interview questions arrived with replies 145–150. The 2026-10-28 interview's Q7–Q9 are archive-lane (see §0) and are carried to the check-in lane, not answered here.
 ## 7. Drafting / tracking + corrections log
 Pipeline tally unchanged: **11 submitted / 2 interviews / 0 offers / 2 rejections.** New queue file `apply-queue-2026-10-29.md` supersedes the 10-28 queue: item 25 urgency line removed, item 32 carries no expiry line, count definition corrected to 33 live / 4 closed (one door per canonical URL, Turing 15), Workday probe row retired, Ashby presence streak resumed, items 33–34 added as possible.
 ## Open for Alexander (via Timmy, Telegram — NOT via Zeph)
 **NEW / CHANGED today:**
 1. **Items 33–34: take or pass** — new verified doors (33: Turing agent-trace $100/hr, Azure-depth caveat; 34: Data Annotation DevOps up to $150+/hr, K8s/Terraform-comfort caveat).
 2. **Item 25 urgency WITHDRAWN** — no deadline attaches to OpenTrain `validThrough` (render-synthesized, build+30d); take-or-pass (security-conditional) stands without an urgency line.
 3. **Turing doors wp680rglf0 + JPQ-cmFZSA now verified** — no longer held as unverified; Turing count 13 → 15.
 4. **micro1 item 5 closure not re-confirmed today** — failed fetch gap, not a verdict change; standing CLOSED verdict untouched.
 5. **Workday probe row retired** — probe role only, not a door.
 **Standing (carried, already re-raised — not re-decided here):** Mercor talent network take-or-pass as a pipeline row (band = past listings, not a promised rate); Surge AI go/no-go + 2-hour screen (pause Mac automations before the screen); workfindy go/no-go (DM-the-poster); Turing N1 go/no-go + rate $100–150; Turing M2 go/no-go + rate; N2 take-or-pass; Askable OSS-gate question + Goleby direct contact; Alignerr N5 take-or-pass; Terac $65/hr window + Baker contact; OpenTrain item 25 take-or-pass (security-conditional, urgency withdrawn above); Inbox items 31/33/34; direct-lane drafts held for his send; Turing doors L2/L3/L4; Alignerr doors L6/L7/L8; Handshake AI trainer; micro1 OSS doors items 13/24 queue-adds (GitHub-gate honest read); items 29/30 (Timmy read: pass, not re-raised); item 31 take-or-pass; item 32 take-or-pass (security-conditional, no expiry line).
 ### Corrections log (replies 145 + interview §2, applied 2026-10-29, append-only)
 - **Superseded: daily-2026-10-28 §1.1 ("validThrough 2026-11-26, 29 days remaining as of today, datePosted 2026-10-07") and §7 / queue item 25 expiry cell ("expiry CORRECTED to validThrough 2026-11-26").** Both are wrong, and the mechanism is Timmy's reply 145 §0: OpenTrain `validThrough` is synthesized at render time as build-time + 30 days. Evidence: item 25 reads 2026-11-23 (10-24, 208,695 B) → 2026-11-26 (10-27, 211,050 B) → 2026-11-27 (10-28, 211,437 B), each exactly 30 days after its own render instant, datePosted byte-identical throughout (2026-10-07T10:19:56.744Z); item 32 in the same minute 10-28 carries the identical 2026-11-27 stamp on a different posting. The "50-day computed span" arithmetic built on the 10-27 value is withdrawn with it. The 10-24 cell (2026-11-23) was right when read as a render stamp and wrong only when treated as a deadline — the error was the treatment, not the reading. **Standing rule, replacing rule (b): OpenTrain `validThrough` is never quoted as a deadline; if it must appear, label it `render-synthesized, build+30d`.** Item 25 keeps its take-or-pass (security-conditional, $150/hr, 20+ h/wk, posted Oct 7) with NO urgency line. Item 32 never receives an expiry line. My 10-29 text fetch of item 25 again shows no validThrough block (extraction drops JSON-LD) — consistent, not a refutation.
 - **Superseded: daily-2026-10-28 §1.8 reason ("text fetch renders full role copy with NO closure language — extractor limitation, inconclusive").** Reply 145 §3: the closure is in a plain fetch (200/213,580 B, `This listing has closed` ×2, `isClosed:true`). The CLOSED verdict itself stands (10-25 render + reply-99 + 10-28 addendum render); only the reason was wrong. **Standing rule: when the extractor is silent, run the second instrument before writing "inconclusive"** — the pay-cell rule, applied to closure markers. The phrase "extractor limitation" is retired for this URL class.
 - **Superseded: daily-2026-10-28 §3 closing line ("Prior-set aggregate unchanged: 13 live / 4 closed").** That 13 was Turing-only, mislabeled as a cross-family total (reply 145 §4). **Going forward: 31 live / 4 closed, one door per canonical URL, Turing 13 of the 31** (interview §2.3). The old line stands as the wrong reading it was.
 - **Q116 close condition:** this file is the landing the amended done-means requires — mechanism named, superseded lines named by file and line, no committed file edited. Item closes on this artifact landing, subject to Timmy's verification.
 ---
 Compiled by Zephyrus Sparkborn, 2026-10-29. Liveness from a clean external network; dated reads, not properties. Failed fetch = failure, never closure. Text fetch ≠ render: a panel or badge a text fetch cannot see is not evidence of absence — and a silence that repeats across instruments is a reason to run the second instrument, not to upgrade the silence into a verdict. Full scout tables: `hidden_files/career-scout-2026-10-29.md`.