Daily career-pipeline tick — 2026-10-28 (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 2026-10-28 ET. No submissions, no accounts, no contact made by me.

0. Pipe sync (this run)

Folded Timmy's replies 129, 130, 131, 132, 133, 134, 135 (2026-10-27), the 2026-10-27 interview, and 136, 137, 138, 139, 140, 141, 142, 143, 144 (2026-10-28). Watermark 128 → 144.

Reply 129 §2 — three career pay cells corrected from Timmy's raw-byte reads: item 29 = $20-100/hr (Seattle 63d40d5a 200/42,042 B and NY 2d057dd9 200/41,857 B, header badge span); item 30 = $50-75/hr (95a64c38 200/37,090 B, three times independently: header badge + meta description + JSON-LD shortDescription); item 28 reproduces $60-120/hr (42,957 B today vs 43,029 B 10-26 — byte drift, figure stable). Item 25: validThrough PRESENT in bytes, reads 2026-11-26 (not 11-23), datePosted 2026-10-07 — a 50-day computed span, not an employer promise. N5 variance CLOSED (bytes half discharged: $60-80/hr in the header span). Terac $65 re-read open (200/224,340 B, Earn panel + isClosed:false). micro1 item 24 spot-check open (zero "closed").
Reply 129 §5 + 130 §3 — two process rules ADOPTED this tick: (a) every new Alignerr/micro1/Terac pay cell gets a raw-byte read with its figure before it is written; "unstated in text fetch" is no longer an accepted cell value for a page of this family. (b) a validThrough value is quoted with the datePosted it sits beside. Reply 130's arithmetic correction applied: "30 days remaining as of 10-27" is remaining time, not span — the span beside datePosted is 50 days. Adopted quote form: "validThrough 2026-11-26, 29 days remaining as of 2026-10-28, datePosted 2026-10-07".
Reply 130 §3 + interview §1.3 — the presented surface: Timmy is NOT re-raising decisions 14/15; his read stands (pass: $20-100 and $50-75 both sit under the $100–150 band). He presents only raw-byte figures to Alexander until this tick file lands with the corrections folded. This file is that landing: queue items 29/30 carry the corrected pay cells, item 25 carries 2026-11-26 + datePosted, N5's variance note closes. No committed file was edited; corrections land here, append-only.
Reply 129 §0 — Timmy's own mistakes, noted not graded: the raw-byte rule was his finding left as a correction instead of a step (now a step, per (a) above); his item-28 byte count printed as an identifier (dated reading); his reply-128 compression mutant description corrected (revert difference, not compression difference).
Replies 131–144 + interview (non-career parts): tooling/control only, no career content, no career asks, no decisions for Alexander. Q60 CLOSED at 33f8f0f (interview §1.1) with a fold condition on my archive lane (verdict tables into the script, two conditions) — archive lane, not this tick. Q96/Q104/Q105 open in other lanes, not mine to take here. Career-lane blobs unchanged across those windows. Repo-visibility line still reads private:false (reply 129 §6) — standing blocker, out of this lane.

1. Time-sensitive / conversion (lead first)

OpenTrain item 25 — LIVE, validThrough 2026-11-26 (page-computed), 29 days remaining as of today, datePosted 2026-10-07. $150/hr ×3 in today's text fetch, posted Oct 7 2026, 20+ h/wk. Security-conditional only — the package on file shows no appsec/SAST/CodeQL depth; ask, don't assume.
Items 29–30 pay CORRECTED (see §0 + corrections log). Item 29 $20-100/hr, item 30 $50-75/hr — both below band; Timmy's read: pass. Decisions 14/15 stand presented, not re-raised.
NEW door item 31 — Turing bAJ2YQwFKG (systems-Python family). Contractor, 10–40 h/wk, 1 month + extensions, US-based; Required Skills: systems/infrastructure/backend in Python/C++/Rust/Go; overview mentions JavaScript (including ReactJS) — React caveat, same class as N2/L4; pay unstated in text fetch; AI video interview required. Take-or-pass; flagged as an L4-family alternate door, counted per the Turing /r/ convention (each /r/ is an application door).
NEW door item 32 — OpenTrain Secure Coding-Agent Benchmark Engineer (M2 twin). Canonical detail page recovered from the OpenTrain cybersecurity category page and verified open today: remote worldwide, 4–8 wks, 20+ h/wk, PT overlap; 8+ yrs + degree line + Python mandatory + Java/Go/C++ + secure-fix experience; pay NOT provided in the listing. Security-conditional take-or-pass.
Item 20 gains a NY variant (not a new item). Alignerr c002014e (New York copy) verified open today, same role body as L6 (Berlin copy): hourly contract, remote, 10–40 h/wk, 5+ yrs. Per the item-29 city-variant rule this is item 20's second variant, not a new numbered item. The scout's "new lead" call on this page is corrected here to variant status — count unchanged.
Workday probe window — LAST DAY TODAY (probe role only, NOT a pipeline door). Today's text fetch TIMED OUT (upstream fetch failure, no HTTP status) — a failed fetch, not a page change. Live-browser recovery was dispatched and is pending at push time; the gate stands on the 10-25/10-26/10-27 measurements (account first, step 1 of 6). The probe row retires after today regardless: the page's own dated line reads "at least until October 28, 2026".
Terac $65 — LIVE. 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 (200/224,340 B) + 10-26 render.
Terac $39 — CLOSED stands. Today's text fetch renders full role copy with NO closure language and NO Apply element — the same inconclusive pattern as 10-27, classified as extractor limitation, not a reopening. The standing verdict rests on the 10-25 live render + reply-99 corroboration. A live-browser re-check was dispatched and is pending at push time; do NOT route until a render speaks.
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, GitHub-gate honest read stands; Turing Xqiv8zPNaw up to $150/hr (security-conditional, 2-week shape).
Turing CUZsKnwt7P — STALE-RISK flag stands. Still rendering, "$75/hour" + "Project start date: May 4, 2026 (Tentative)", rolling "until the batch is full". Weigh as backfill/batch, not a fresh project.

UNVERIFIED (hold, do not queue): Handshake individual listings (hub renders, list doesn't); Turing wp680rglf0 + JPQ-cmFZSA alternate doors (scout-reported, not my-read — corroboration-requested); micro1 Senior Platform Engineer re-post (mirror-only, no canonical URL). CLOSED: micro1 items 5 and 10 (markers re-confirmed verbatim today); Terac item 12 ($39 door, standing verdict — see §1.8). WITHDRAWN (not closed — removed at source, never a door): item 27 Gramian (not re-queued). Dropped, never a verified door: Alignerr beae0b1f-… (Unrecovered).

2. ATS gate census — dated re-verification (2026-10-28)

Platform	Reference	Today's read
Greenhouse	Point72 `8356788002` (same probe)	Direct open again returns header/footer only ("Your submission is processing") — form fields remain search-capture-based, as flagged 10-24→10-27. 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)	Posting header renders (Security Engineer, Cloud; NY HQ; $211.4K–$290.6K; published 2026-04-07) but the form body returned "Fetching application form" only — the field set was NOT re-fetched today. Code-analysis question presence NOT re-confirmed today: the 10-25/10-26/10-27 streak pauses on an instrument limitation (form shell only), NOT a removal. No post-submit emailed code or segmented code widget visible.
Lever	Worldscape Technology `6b5089a8-2daf-4b4e-a739-bcf762cfe29a/apply` (probe role only)	Field set re-fetched in full, identical to 10-24→10-27 (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 element NOT visible in this static text — consistent with reply 100 §5(a): the captcha lives on the submit path, which static form text does not surface. Gate line stands as corrected: LinkedIn auto-complete + candidate attestations + hCaptcha-on-submit*. Today's fetch does not re-verify the captcha and does not contradict it.
Workday	NVIDIA `JR1997186` (probe role only)	Text fetch timed out today (upstream fetch failure, no HTTP status). Live-browser recovery dispatched, pending at push time. Gate stands on 10-25/10-26/10-27 measurements: account first, "Create Account/Sign In" step 1 of 6. Probe window reaches its last day today per the page's own dated line ("at least until October 28, 2026"); the row retires after today regardless of today's re-confirmation. Census probe only, not 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.

Gate verdicts carried, no change: Greenhouse verification-code wall (Timmy lane, post-submit); Lever hCaptcha-on-submit (reply 100 §5a, corrected line); Workday account gate (live-measured 10-25, re-confirmed 10-26 and 10-27 by both instruments; today timed out, recovery pending); Ashby spam-flag risk + code-analysis question (presence-confirmed 10-25→10-27, NOT re-confirmed today — form shell only); Niceboard per-board route (vendor-documented, no uniform form).

3. Open-application status

Re-read 2026-10-28 (text fetches for all tracked URLs; live browser pending for Workday + Terac $39):

Turing 12/12 LIVE → 13 doors with item 31, 0 title/pay changes across the 12: 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.
micro1: items 13 and 24 re-read OPEN (13: $100–150, published 2026-09-16, GitHub-gate stands; 24: $150–300, "Apply now" renders, published 2026-08-14); items 5 and 10 closure markers re-confirmed verbatim ("This job is currently closed and not accepting applications.").
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 2026-10-05T16:44:58Z, "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 the 10-26 render + Timmy raw-byte read (§1.7). Text fetch still shows "Hourly Remote" only — the pay lives in the side panel the extractor drops, not in the header or body.
Terac $39/hr — CLOSED (standing verdict, §1.8). Today's text fetch shows role copy with no closure language and no Apply element — same extractor limitation as 10-27; the closure stands on the 10-25 live render + reply-99 corroboration. Do not route.
Alignerr ×8 pages LIVE, unchanged, + item 20 NY variant verified: N5 pay ($60–80/hr) not re-shown in text fetch — body cell now carries Timmy's raw-byte Documented read (variance note CLOSED, §0); item 28 pay ($60–120/hr) not re-shown in text fetch — cell carries Timmy's raw-byte read; items 29 ($20–100/hr) + 30 ($50–75/hr) corrected per §0 (Timmy raw-byte reads; my extractions do not surface Alignerr header badges, same class as N5/Terac $65); L6 Berlin-copy vs remote-terms nuance stands; L7, L8 unchanged.
Handshake — hub LIVE, listings UNVERIFIED (hub shell + marketing only; individual opportunities did not render).
OpenTrain item 25 — LIVE (see §1.1 for the corrected expiry note).
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 (see §1).

Prior-set aggregate unchanged: 13 live / 4 closed / 0 re-posted, plus item 28 (14th live door, pay-corrected), items 29–30 (possible, now pay-corrected, Timmy read: pass), and new possible items 31–32 (pending Timmy fold). Turing door count moves 12 → 13 with item 31.

4. New gig leads (verified open 2026-10-28)

Full scout tables: hidden_files/career-scout-2026-10-28.md.

#	Role / URL	Terms	Pay	Fit
31 (possible)	Turing — Senior Python Engineer – LLM Evaluation (US-based) — https://work.turing.com/r/bAJ2YQwFKG (canonical page fetched open; L4-family alternate door)	Contractor, remote US-based, 10–40 h/wk, 1 month + extensions; AI video interview required	Not stated in text fetch	3+ yrs; systems/infrastructure/backend in Python/C++/Rust/Go. Overview mentions JavaScript (including ReactJS) — React caveat, same class as N2/L4; do not claim React. Take-or-pass.
32 (possible)	OpenTrain — Secure Coding-Agent Benchmark Engineer — https://www.opentrain.ai/jobs/secure-software-coding-agent-benchmark-engineer--cmv0u6y87000k06i0ytsnui5a/ (canonical detail page recovered from the cybersecurity category page, fetched open)	Remote worldwide, contract part-time, 4–8 wks, 20+ h/wk, PT overlap	Not provided in the listing	M2 twin shape: 8+ yrs + degree line + Python mandatory + Java/Go/C++ + secure-fix experience. Security-conditional take-or-pass.
20-variant	Alignerr — Senior SWE — Agentic Coding, New York variant — https://www.alignerr.com/jobs/c002014e-c0a5-4a17-958c-6e2e4169ce71 (canonical page fetched open)	Hourly contract, remote, 10–40 h/wk	Not surfaced in text fetch (header-badge class)	Same role body as item 20 / L6 (Berlin copy). NOT a new item — second variant of item 20 per the city-variant rule.

Screened out today: Turing mWC6JZJxcH (US excluded, re-confirmed); Turing URNPQo66KW (React required in Required Skills); Turing 1WxyUT1Gif (top-tier-company pedigree gate + ReactJS throughout); Alignerr Ruby a62a41c6 (stack mismatch); OpenTrain Application Security AI Evaluation Expert $40–80/hr (below band at low end + appsec specialty); micro1 Senior Platform Engineer re-post (mirror-only, no canonical URL; also AWS/K8s-adjacent fit problem); OpenTrain related cards (ML Researcher / ML Applied Scientist $60–120/hr equiv — related-card text only, no new canonical detail verified; earlier ML Researcher degree-gate screen stands).

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 — Series A Dec 2024, corrected); Zachary "Zac" Baker (Terac Co-founder & CEO — contact for the surviving Terac $65 door; the $39 door is closed); 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 interview questions arrived with replies 129–144 (the 2026-10-27 interview's questions were archive-lane Q4–Q6, answered in the check-in lane, not here). Nothing owed on the career lane from the interview file beyond §1.3's presented-surface rule, which this file satisfies by landing the corrections.

7. Drafting / tracking

Standing drafts suffice (Timmy's item-3/4/5 paste-ready pitches from 10-14; candidate-answers-v1.md; 24-MAG pitch — verbatim-applicable to the Turing door, same project). Pipeline tally unchanged: 11 submitted / 2 interviews / 0 offers / 2 rejections. New queue file apply-queue-2026-10-28.md supersedes the 10-27 queue: items 29/30 pay corrected, item 25 expiry corrected, N5 variance closed, items 31/32 added as possible, item 20 NY variant noted, Workday probe row retired after today, Ashby presence streak paused (not removed).

Open for Alexander (via Timmy, Telegram — NOT via Zeph)

NEW / CHANGED today:

Items 31–32: take or pass — new verified doors (31: Turing systems-Python door, React caveat, pay unstated; 32: OpenTrain security-benchmark door, M2 twin, pay unstated, security-conditional).
Items 29/30 pay CORRECTED ($20-100 / $50-75) — decisions 14/15 stand presented with Timmy's read (pass); not re-raised.
Item 25 expiry CORRECTED to validThrough 2026-11-26, 29 days remaining as of today, datePosted 2026-10-07 (page-computed) — security-conditional take-or-pass stands.
Item 20 NY variant verified — same role as L6, not a new decision.
Workday probe window ends today — probe role only, not a door; the census row retires after today.

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, expiry corrected 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).

Corrections log (replies 129/130, applied 2026-10-28)

Append-only; the original lines stand, corrected beside them.

Reply 129 §2.1 — Item 29 pay corrected. The 10-27 daily §1.3 / queue item 29 "pay unstated in text fetch" is wrong: both canonical pages state $20-100/hr in the header badge span (Timmy raw-byte reads 2026-10-27, 200/42,042 B Seattle + 200/41,857 B NY; re-verified reply 130 §3 at identical bytes). Decision 14 reworded accordingly. Provenance caveat, carried honestly: my 10-28 browser.open extraction again surfaces NO pay line on either page — the header badge is dropped by extraction, same class as item 28 / N5 / Terac $65. The cell is therefore Timmy-Documented, not mine; it is not corroborated on my instrument and is not refuted by the extractions.
Reply 129 §2.2 — Item 30 pay corrected. The 10-27 "pay unstated in text fetch" is wrong: the canonical page states $50-75/hr three times independently in the bytes (header badge + meta description + JSON-LD shortDescription; Timmy raw-byte 200/37,090 B, re-verified reply 130 §3 at 200/37,170 B). Decision 15 reworded accordingly. Same provenance caveat as item 29: my extraction does not surface it.
Reply 129 §2.3 — Item 28 cell kept, byte count dropped. $60-120/hr reproduces (42,957 B on 10-27 vs 43,029 B on 10-26 — dated readings, not identifiers). The queue cell keeps the figure and drops the notional byte count.
Reply 129 §2.4 + 130 §3 — Item 25 expiry corrected. The standing 2026-11-23 cell does not reproduce: the page's JSON-LD reads validThrough 2026-11-26 beside datePosted 2026-10-07 (Timmy raw-byte 200/211,050 B). That is a 50-day computed span with 29 days remaining as of 2026-10-28 — quoted in that form from here on, never as a bare expiry date, and never as an employer promise. My text fetches (10-27, 10-28) again do not surface validThrough — the JSON-LD block is dropped by extraction, same class as the pay badges.
Reply 129 §2.5 — N5 variance note CLOSED. $60-80/hr now stands Documented from Timmy's raw-byte read (header span), not only from the structured-data read. Nothing about the door changes.
Reply 129 §5 / 130 §3 — Process rules adopted (standing): (a) raw-byte pay read before writing any new Alignerr/micro1/Terac pay cell; (b) validThrough always quoted beside datePosted.
Scout correction (this tick, mine): the scout's Alignerr c002014e "new lead" is re-graded as item 20's NY variant (same role body as L6 Berlin copy) — not a new numbered item, per the item-29 city-variant rule.

Compiled by Zephyrus Sparkborn, 2026-10-28. 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-2026-10-28.md. Pending at push time: live-browser recovery for Workday JR1997186 (text fetch timed out) and Terac $39 render re-check — both recorded here as pending, neither as a verdict.

21 KiB Raw Permalink Blame History Unescape Escape