Process Newsletter — project README

Goal

Ingest email newsletters from a whitelisted set of senders into the vault as structured assessment notes, with bias and sponsor flagging. Separate from Substack-specific tooling — this works on any newsletter that lands in ben@raydata.co regardless of publisher (Substack, Ghost, Mailchimp, self-hosted).

Companion skill: ~/.claude/skills/process-newsletter/SKILL.md.

Status

• Prototype complete as of 2026-04-11 using SeattleDataGuy's 9-email history as the proving ground (8 backfilled, 1 was pre-existing).
• Skill file drafted. Whitelist locked in.
• Remaining K senders not yet backfilled (pending founder go-ahead for the next batch).

Whitelist — locked in 2026-04-11

K = keep, backfill history + ongoing watch

⚠️ Count-before-budget rule (added 2026-04-30)

The original "N+ in history" estimates in the K table were Substack-archive-based, NOT Gmail-inbox-bounded. Real inbox counts have come back substantially smaller (Stratechery: 60 vs 201+ estimate; PDM: 32 vs 20+ estimate; Not Boring: 28 vs 201+ estimate). The gap is because Gmail only has from-subscription-date forward, not the sender's full archive.

Rule for remaining un-scanned senders (Every, Commoncog, Analytics Engineering Roundup, ARK Invest, Write With AI, Ship30for30, Essay Architecture): run discovery with --dry-run semantics first — count messages in inbox BEFORE planning batch sizes. If count is small (<20), skip the batch-task overhead and process inline via Mode 3 (Backfill, legacy small-sender path). Don't allocate 10+ batch tasks for a sender that only has 25 messages in the inbox.

Implication: the total backfill work is significantly smaller than the README originally implied. Prioritize the per-sender deep-fetch quality (sponsor detection, RDCO mapping discipline) over volume planning.

F = follow-forward only, no backfill (watch from now onward)

Known sender-specific gotchas (learned from SDG backfill)

Innermost Loop — timestamp vs content-date offset (added 2026-08-06)

• Issues are titled "Welcome to " but arrive late-night for the next calendar day — e.g. the email with subject "Welcome to August 5, 2026" lands 2026-08-06T01:27Z. Watch-mode date matching by send-timestamp alone will flag these as new even when already filed. Always confirm against source_url in the candidate existing file before treating an Innermost Loop thread as unprocessed.
• 2026-08-25 confirmed miss: the parent dispatched the "Welcome to August 25" thread as new because its topic slug (fable5/singularity-repricing) looked different from the already-filed 2026-08-25-innermost-loop-fable5-share-loss-thread-lines-autonomous-weapons.md. It was an exact duplicate — same source_url, same full item list. A digest's slug names only 2-3 of its many threads, so slug mismatch alone is not evidence of a different issue. Always check source_url (not just topic-slug vibes) against existing files before dispatching an Innermost Loop message as unprocessed.
• The ship30for30 "ChatGPT Work for Writing Businesses" bootcamp campaign (Aug 2026) ran a multi-email sales sequence (quiz → career-motivation listicle → bonus-stack close) around the original technique email (2026-07-29-ship30for30-chatgpt-work-lead-generation.md). All three downstream emails triaged as pure sales with zero new extractable content — confirms the "read first 200 words" rule catches campaign sequences correctly.

SDG — SeattleDataGuy (fully backfilled)

• Sponsor pattern: Estuary (disclosed adviser relationship) shows up in the top-of-article pre-amble ~60% of the time. SDG also switches to explicit self-consulting CTAs ("today's article is sponsored by me, the Seattle Data Guy!") — sometimes at top, sometimes mid, sometimes at bottom. Skill should flag both forms and not key detection off placement. Added 2026-05-14: Greybeam appeared as a third-party paid sponsor in the "What Leading a Data Team Actually Looks Like Right Now" issue — first new SDG sponsor observed since Estuary. Watch for whether this is a one-off or a new recurring relationship.
• Curation section ("Articles Worth Reading") is roughly 50/50 self-cross-promo vs third-party. He frequently links his own prior posts in the curation slot. The skill should label links by comparing the linked domain/author to the newsletter sender.
• Series structure: SDG runs multi-month series (2026 "data pipelines" series, currently 6 articles in). Later articles reference earlier vocabulary. Backfill order matters — go oldest-first so cross-references make sense.
• Guest posts exist — Olga Berezovsky hosted on Jan 23. Guest posts come with cross-promotion for the guest's own newsletter in the top slot (different from the Estuary pattern).

PDM — Practical Data Modeling (Joe Reis) — partial backfill, paid sub lapsed 2026-04-24

• Substack expiry sequence (3-email cadence): "ending in a week" → "ending tomorrow" → "subscription ended." All three are pure billing/admin emails with zero extractable content. Skip on subject-line pattern alone in Mode 4 watch — don't read the body. The 2026-04-30 PDM batch [1-5] confirmed this template.
• Pulse-survey launch vs results mislabel risk: A "New pulse survey just dropped" email is the LAUNCH (CTA to take the survey), NOT the results. Joe Reis's pattern is to present findings at a keynote (e.g., Stockholm May 7 2026) and publish a results-with-dataset post afterward. When discovery pre-triages a "pulse survey" subject, default the assumption to "launch CTA" and look for the follow-up results post 2-4 weeks later. The results post is the high-signal artifact; the launch is borderline-skip with one extractable data point at most.
• Series structure: Long-running MMA chapter series (Ch 1-16 already filed, Feb-Apr 2026). Cross-references rely on chapter order — process oldest-first when backfilling.
• Paid-tier risk on results posts: Reis ships results-with-dataset posts as paid-tier content. Watch the post-Stockholm May 7+ window — if plain_text returns paywall stub, founder needs to resubscribe (lapsed 2026-04-24) before that batch can process the substantive body.

Not Boring (Packy McCormick) — partial backfill, ongoing watch

• Hybrid cadence: Mon long-form essay + Fri "Weekly Dose of Optimism" curation. Both formats live in the same inbox; classify per-issue, not per-sender.
• Co-written essays = founder pitches. ~4 of 11 historical long-form essays follow the "Packy hands the keyboard to a portfolio-adjacent founder" pattern (Venezuela, Great Blue Frontier, etc.). Treat the company list and "what's missing in the platform map" framing as the co-author's go-to-market collateral, not neutral analysis. The underlying framework is usually still useful; the company endorsements are not. Frontmatter should byline both authors AND surface the structural bias in a Sponsorship section even when no explicit sponsor block exists. The Apr 23 2026 Great Blue Frontier essay (Will O'Brien / Ulysses Maritime) is the canonical exemplar — Packy explicitly disclaimed Not Boring Capital is NOT an investor, but the entire essay is a Series-A-adjacent positioning piece.
• Two layers of sponsor on co-written essays: (a) explicit paid sponsor block (Framer, SVB, etc.) — clean disclosure, low bias; (b) structural sponsor — the co-author's company. Always disclose both, separately.
• WDoO portfolio-disclosure ambiguity: Packy doesn't always disclose his angel positions in the curation slot. Absence of "Not Boring Capital is an investor" caveat ≠ no relationship. When a curation item is enthusiastic about a specific company (Apr 24 2026 Medra was the trigger), do a quick public-investor-list verification before treating the framing as neutral. Note the verification result in the assessment ("verified Not Boring Capital not in public Series A list" or "could not verify, treat as messenger").
• Sister-author cross-promo in WDoO: Kevin Kwok essays appear in the curation slot ("first Kevin Kwok essay in nearly a year") — friendly-network amplification, not strict third-party. Flag but don't disqualify; the analysis is usually substantive.
• Series structure in WDoO: themes recur week-to-week (geothermal Quaise → Fervo S-1; AI-for-bio bottleneck → Medra throughput answer). Cross-link adjacent WDoO entries.
• Inbox vs archive count gap: Original "201+ in history" estimate was Substack-archive-based. Real Gmail inbox has ~28 messages from-subscription-date forward. Confirmed pattern across multiple senders.
• Off-cycle reactive posts can be paywall-truncated (new 2026-07-25): Packy occasionally ships a short reactive post outside the Mon-essay/Fri-WDoO cadence (e.g. "AI is Oil, Not God," a same-day reaction to a news event). These can cut off at the free-preview boundary mid-argument, with the substantive rebuttal/analysis behind the subscriber wall. File with source_fidelity: paywall-truncated and say so explicitly in the Why-in-vault section — don't present the free teaser as the full argument.
• Reducto sponsor — CONFIRMED standing relationship (updated 2026-09-18): Reducto sponsored WDoO #209 (2026-09-04), #210 (2026-09-11), AND #211 (2026-09-18) — three consecutive issues, clearing the confirmation bar used for Technically/Railway and CFO Secrets/Summation. Treat Reducto as WDoO's default sponsor going forward until a future issue breaks the pattern.
• Reducto streak broke at #212 (2026-09-25): sponsor was Jack & Jill (recruiting matchmaking platform), not Reducto — clean ad block, no investor disclosure. Three consecutive issues was not a permanent standing relationship; revert to unknown-until-scanned per issue for WDoO sponsors, same posture as AlphaSignal/CFO Secrets/Mostly Metrics.

CFO Secrets — "Inheriting a Shitshow Finance Function" series CLOSED at 4/4 Stuut (added 2026-09-26)

• Part IV ("Gutting the finance team," 2026-09-26) confirmed Stuut as sponsor for the fourth and final installment, matching Parts I-III. Series-sponsor pattern now fully closed, not just probable: Stuut was the de facto standing sponsor for all 4 parts of this mini-series, same confidence tier as Summation/"Building FP&A" and Summation/Man-Utd. No further installments expected — nothing to watch going forward for this specific series.
• Filename-slug note: this issue was filed as 2026-09-26-secret-cfo-gutting-the-finance-team-iv.md. Both cfo-secrets- and secret-cfo- prefixes have been used interchangeably across the whole backfill (confirmed via ls ~/rdco-vault/06-reference/, e.g. 2026-06-23, 2026-07-25, 2026-08-08/22/25/29, 2026-09-01/22 all use secret-cfo-) — this is pre-existing inconsistency, not a new defect. Cross-links still resolve fine via full wikilink regardless of prefix.

CFO Secrets — sponsor pool 11th member (added 2026-09-22)

• Numeric (numeric.io, "data-centric accounting" webinar) appeared as a clean paid UTM block, no author relationship, in the 2026-09-22 Tue Mailbag ("CEO with a secret salary..."). Confirmed rotating-pool member #11: Campfire, Zip, Pulley, Stuut, Una, Ledge, Summation, Nominal, CloudZero, Aleph, Numeric. Logging future new entrants here going forward rather than appending to the K-table cell, which is already ~5KB and read in full every run.

ARK Invest — Gmail body is teaser-only, source site Cloudflare-blocked (added 2026-09-22)

• The 2026-09-22 "A Letter To Investors" issue rendered in Gmail as a pure teaser (opening line only, no argument body). WebFetch/curl against ark-funds.com / arkinvest.com hit a Cloudflare challenge page, not the article. Recovery path: WebSearch for ARK's related public commentary (their "In The Know" series covers the same theses) plus third-party coverage of the same argument. Filed with source_fidelity: reconstructed-from-web. Treat ARK Invest source URLs as generally unfetchable directly — plan on the WebSearch fallback by default rather than trying WebFetch first.

Every — sponsor observation (added 2026-08-20)

• First observed sponsor: Brief (briefhq.ai, prototyping/spec tool) — clean paid block plus a house "Want to sponsor Every?" CTA, in the "In Defense of AI Writing" Context Window issue. One-off so far — watch whether it recurs or rotates like other senders' sponsor pools.
• 2026-08-24 check ("I Tried the AI Model Built to Fix AI Writing," Working Overtime, Katie Parrott): no paid sponsor block — only a research-disclosure line ("Deft had no input on the development of this review") plus the standard house footer (Every All Access, Sparkle/Cora/Spiral/Monologue cross-promo). Confirms Brief remains a one-off, not yet a recurring relationship.

Patterns to watch for in other senders (not yet backfilled)

• Stratechery: known to have "This Week in Stratechery" weekly digest + interview-format issues + daily updates. Likely a mix of formats per sender. Needs format-detection per-issue, not per-sender.
• Every: multi-author publication, each issue may have a different byline. Treat each issue as its own author.
• ARK Invest: pure investment commentary. Track for bias since it's a fund promoting its own positions.
• Ship30for30 / Essay Architecture: writing craft — useful meta, but heavy on marketing-style CTAs. Sponsor detection will be noisy.
• Data Elixir: curation-only once it arrives. Watch for bias toward the tools/platforms Lon is sponsored by.

AlphaSignal — sponsor pair broke after 2 days (2026-09-02 → 2026-09-03)

• Unblocked + Datadog appeared as the two "Presented by" partner blocks on both the 2026-09-01 (2026-09-01-alphasignal-timesfm3-claude51-runway-solaris) and 2026-09-02 (2026-09-02-alphasignal-claude-fable51-cursorbench-agent-software) issues — identical pair, two consecutive days, initially flagged as a possible standing rotation slot.
• 2026-09-03 (2026-09-03-alphasignal-claude-commerce-agents-qwen3-gemini25flash) broke the pattern: sponsors were OpenRouter, Vanta, Granola — none matching Unblocked or Datadog. Two consecutive days was a short-run coincidence, not a standing rotation. Note also that OpenRouter and Granola themselves recurred from the 2026-09-01 alphasignal-claude-code-limits-grok-shopping issue (OpenRouter, Teleport, Granola) — AlphaSignal appears to draw from a broader rotating pool (OpenRouter, Teleport, Granola, Unblocked, Datadog, Vanta, ...) rather than a fixed daily pair. Treat AlphaSignal sponsors as unknown-until-scanned per issue, same posture as CFO Secrets/Mostly Metrics.
• Also noted 2026-09-03: an unlabeled masthead "In Partnership with" logo slot at the top of the issue, separate from the per-item "Presented by" blocks, rendering as an image with no legible alt text or identifiable link target — worth checking future issues for whether this masthead slot is a distinct, identifiable sponsor once image content can be resolved.
• 2026-09-04 (2026-09-04-alphasignal-cursor-self-hosted-agents-claude-shopping-dlss5): three MORE distinct paid placements — Finest (AI-spend cost router), Google Cloud (GPU/TPU capacity playbook), Tiger Data (Postgres metrics/embeddings/analytics store, appeared inside the Signals list rather than a standalone block). None overlap with 09-01/09-02 (Unblocked, Datadog) or 09-03 (OpenRouter, Vanta, Granola). Rotating pool now confirmed at 9+ distinct sponsors across 4 days: OpenRouter, Teleport, Granola, Unblocked, Datadog, Vanta, Finest, Google Cloud, Tiger Data. The unlabeled masthead "In Partnership with" slot also recurred, still unresolved.
• 2026-09-07 (2026-09-07-alphasignal-lean-proof-swarm-cheat-whistleblower): Teleport and Google Cloud both recurred (confirmed pool members, not new). Redis (agent long-term-memory pitch, native ad inside the Signals list) is a TENTH distinct sponsor, first seen this issue. Rotating pool now 10+ confirmed distinct sponsors: OpenRouter, Teleport, Granola, Unblocked, Datadog, Vanta, Finest, Google Cloud, Tiger Data, Redis. The unlabeled masthead "In Partnership with" slot recurred yet again, still unresolved.
• 2026-09-08 (2026-09-08-alphasignal-meta-aira3-kaggle-gold-agentic-benchmarks): Google Cloud recurred a third time (confirmed pool member). Two MORE new distinct sponsors: Attio ("first agentic CRM" pitch, standalone "Presented by" block) and Flint AI (Switch — bring any agent into Slack/Teams/Discord, native ad inside the Signals list). Rotating pool now 12 confirmed distinct sponsors: OpenRouter, Teleport, Granola, Unblocked, Datadog, Vanta, Finest, Google Cloud, Tiger Data, Redis, Attio, Flint AI — still growing at roughly one-to-two new entrants per issue, no sign of the pool closing. The unlabeled masthead "In Partnership with" slot recurred again; still unresolved from plaintext extraction (renders as "In Partnership with ·" immediately before "Today's Author," with no name or link in between) — resolving it would require pulling the HTML/image render rather than PLAIN_TEXT messageFormat.
• 2026-09-09 (2026-09-09-alphasignal-navier-stokes-agent-swarm-mistral-funding): masthead "In Partnership with" slot resolved for the first time via FULL_CONTENT messageFormat — QA.tech. Four more new distinct sponsors this issue (QA.tech, Ory, Launch Darkly, Voices), none overlapping the prior pool. Rotating pool now 16+ confirmed distinct sponsors, still growing with no sign of closing. Worth confirming next issue whether the masthead slot stays legible with FULL_CONTENT or was a one-off render quirk.
• 2026-09-13 (2026-09-13-alphasignal-deepseek-v4.1-flash-kv-cache-efficiency): masthead "In Partnership with" slot was ABSENT entirely — the first issue since the 09-09 resolution to carry none. Sponsor was Datadog (State of AI Engineering report), an already-confirmed pool member, appearing as a standalone "Presented by" block instead. Confirms the masthead slot is itself intermittent, not a fixed per-issue feature — don't treat a missing masthead as a scan failure.
• 2026-09-14 (2026-09-14-alphasignal-openai-prompt-bloat-voicestudio-oss-clone): two MORE new distinct sponsors — Span and Wiz — plus Ory recurring (confirmed pool member). Rotating pool now 18+ confirmed distinct sponsors, still growing with no sign of closing. Masthead slot was present but unresolved even under FULL_CONTENT — the image alt text was a generic placeholder (partner_image), not a legible name like 2026-09-09's QA.tech resolution. Masthead resolvability is inconsistent per-issue, not a fixed property of FULL_CONTENT messageFormat — don't assume 09-09's resolution generalizes.
• 2026-09-15 (2026-09-15-alphasignal-flybrain-robot-bytedance-self-improving-rlt): ONE more new distinct sponsor — WorkOS — plus Google Cloud and Attio recurring (confirmed pool members). Rotating pool now 19+ confirmed distinct sponsors, still growing with no sign of closing. Masthead "In Partnership with" slot present again but unresolved (generic alt text) — consistent with the 09-14 pattern, not the 09-09 QA.tech resolution.
• 2026-09-23 (2026-09-23-alphasignal-opus55-gpt6-sol-luna-pricing): THREE more new distinct sponsors — Encord (standalone block, Long Horizon Physical AI Summit), AI Conference (standalone block, event discount), and Nyra (native ad embedded in the Signals list, styled as a "Nyra, 1.4k stars" caption rather than a full block). None overlap the prior 19+ pool. Rotating pool now 22+ confirmed distinct sponsors, still growing with no sign of closing. Masthead "In Partnership with" slot present but unresolved even under FULL_CONTENT (generic partner_image alt text) — consistent with the 09-14/09-15 pattern, not the 09-09 QA.tech resolution.
• 2026-09-24 (2026-09-24-alphasignal-950-agent-enzyme-discovery-claude-code-cloud): ONE more new distinct sponsor — pre.dev (native ad in the Signals list) — plus Wiz and Google Cloud recurring (confirmed pool members). Rotating pool now 23+ confirmed distinct sponsors, still growing with no sign of closing. Masthead "In Partnership with" slot present but unresolved (generic partner_image alt text), consistent with the recent-run pattern.
• 2026-09-25 (2026-09-25-alphasignal-claude-marketplace-unified-billing-llm-fingerprinting): THREE more new distinct sponsors in a single issue — Prior Labs (TabPFN-3.5 benchmark pitch, standalone block), Origin Technology (agent runtime observability, standalone block), and Sentry (native ad in the Signals list) — the fastest single-issue pool growth logged to date. Rotating pool now 26+ confirmed distinct sponsors. Masthead "In Partnership with" slot cleanly resolved to Voices (recurring pool member since 09-09) with full webinar copy — only the second clean masthead resolution since 09-09's QA.tech, most issues in between rendered generic partner_image alt text. Also notable: the issue's own Top News lead story is Anthropic's Claude Marketplace launch (2,000+ plugins, unified billing) — confirmed pure editorial coverage, no sponsor block on that item and none of the named marketplace vendors (Cursor, CrowdStrike, Snowflake, Accenture, Deloitte) appear as paid sponsors elsewhere in the issue.

Every — live-session/Camp promo triage (added 2026-09-04)

• "Today at N ET: " emails (e.g. 2026-09-04 "How we're using Fable 5.1 and GPT-6 Astra") are subscription-upsell event CTAs ("Camps are included with a subscription. Upgrade and you're in.") with no extractable technique in the body — triage-skip on subject+snippet alone, same bucket as the sales-funnel rule. Distinct from Every's regular Context Window / Working Overtime long-form issues, which still process normally.

Data Engineering Central — Delta Lake standing sponsor (added 2026-09-14)

• Delta Lake is a confirmed standing recurring sponsor, not a rotating pool — logged across at least nine issues now (2026-04-22, 05-09, 05-18, 06-11, 07-06, 07-20, 08-10, 08-31, 09-14), DEC's most consistent advertiser to date. Pattern: explicit "Thanks to Delta for sponsoring this newsletter!" block mid-issue plus the author's own disclosed daily-user framing ("I use Delta Lake daily..."). Treat any Delta Lake mention in this newsletter as commercially motivated by default; don't treat it as neutral technical commentary. This is a different sponsorship model than AlphaSignal/CFO Secrets/Mostly Metrics' rotating pools — no unknown-until-scanned assumption needed here, Delta Lake is the default expectation.

Technically — sponsor observation (added 2026-08-11)

• First observed sponsor: Railway (series-level ad block plus in-narrative product placement embedded in the walkthrough example, "Software Eng for Vibe Coders: On Frontends + Backends" issue). One-off so far — watch whether it recurs or rotates like CFO Secrets/Mostly Metrics. 2026-08-18 update: "What's going on with open-weight models?" (Will Raphaelson) carried no sponsor block — Railway did not recur. Treat Technically as unsponsored-by-default until a pattern re-establishes.
• 2026-09-08 update — pattern confirmed at series level, not sender level. Railway sponsored Part 2 (Databases + Storage, Aug 25) and Part 3 (Auth + Security, Sep 8) as well as Part 1. The Aug 18 no-sponsor issue is a different Technically series/author (Will Raphaelson, open-weight models) — it was never evidence against Railway recurring, because it isn't the same series. Railway is the fixed sponsor for the entire 6-part "Software Eng for Vibe Coders" series; the byline also rotates within the series (Justin Gage Part 1, David Krevitt Parts 2-3) without affecting sponsorship. Keep "unsponsored-by-default" for Technically's other series (open-weight models, harness-engineering, etc.), but flag every future "Software Eng for Vibe Coders" installment as Railway-sponsored unless the issue explicitly says otherwise.

Skill design — key decisions

1. Skill over cron for now. The /process-newsletter watch invocation is user-triggered, not scheduled, until we have confidence it doesn't burn context or generate noise.
2. Always-flag, never-filter. Sponsors and bias get disclosed in frontmatter and body, never used to skip an article. The reader needs to see the angle.
3. Deep-fetch cautiously. Max 2 link follows per curation issue, only for third-party links clearly relevant to RDCO topics. No paywall traversal.
4. Vault-path discipline. All notes under 06-reference/<YYYY-MM-DD>-<sender-slug>-<topic-slug>.md. No exceptions. The filename convention is what lets the skill detect "already-filed" without duplicating work.
5. Tracked authors feed Task #4. When a guest post or curation link surfaces a new author worth following (Dylan Anderson from the Mar 25 SDG issue, Olga Berezovsky from Jan 23), they go to the CRM candidate list, not directly into a contact file.

Lessons from the SDG prototype

What worked well:

• Oldest-first order let me notice the series structure and forward-link properly.
• Per-article "Mapping against Ray Data Co" section forced me to justify why I was filing. Several articles passed the mapping bar weakly — which is a useful signal that the discipline is tight, not loose.
• Sponsor disclosure as a top-level frontmatter field means future filtering/reporting is trivial.
• The "This one matters most" framing on the Mar 25 "You Will Know Nothing" article surfaced a real strategic tension for RDCO that I don't think I'd have articulated without being forced to file the note.

What to improve for the next sender:

• Writing 8 articles in one session is context-expensive. Next time: batch fetches, write notes in parallel where possible, cap at 4 articles per session.
• Build a sender-slug lookup (just a dict mapping email → folder prefix) so filename generation is deterministic. SDG slug = seattle-data-guy, Stratechery slug = stratechery-ben-thompson, etc. Confirmed slug for Mostly Metrics is mostlymetrics (no hyphen) — a 2026-08-03 watch run dispatched a sub-agent with the slug mostly-metrics, which missed the already-filed check and wrote a duplicate note before the parent caught and deleted it. Always grep 06-reference/ for the sender's existing files before assuming a slug rather than hardcoding one in a dispatch prompt.
• Consider whether the 06-reference/ folder will get unwieldy as the backfill grows. At ~10 K senders × ~200 messages each = ~2000 files. May need a per-author subfolder once we're past ~50 total.
• The Gmail API's resultSizeEstimate caps at ~200, so "how big is the backfill" requires actual pagination. Not a blocker; just know it.

Next actions (pending founder approval)

1. Backfill the next K sender — recommend Stratechery (201+ messages, highest expected signal density, single author for consistent assessment voice).
2. Set up the watch cron loop once we've done 2-3 senders and trust the pattern.
3. Revisit the 06-reference/ folder structure question at ~50 files.
4. Handle Data Elixir once founder re-subscribes.

Related

• [[../../06-reference/2026-01-05-seattle-data-guy-data-pipeline-patterns]] — first backfilled article
• [[../../06-reference/2026-03-25-seattle-data-guy-know-nothing-and-be-happy]] — the most strategically relevant of the eight
• Task #38 in the task list — "Build /process-newsletter skill (multi-source)" — in_progress
• Task #4 — "Add Twitter users to CRM workflow" — where newly-surfaced tracked authors feed

AlphaSignal — "Sunday Deep Dive" single-thread format observed (added 2026-09-27)

• 2026-09-27 issue (2026-09-27-alphasignal-system-one-models-jev-laya-clm8b.md) was a "Sunday Deep Dive" — one long-form single-thread article, not the usual multi-item Signals curation. Classified single-thread-deep-dive, not curation. Confirms AlphaSignal runs at least two structurally distinct formats depending on day-of-week; classify per-issue from actual structure, don't assume curation by default for this sender going forward.
• Content note: the deep-dive covered TypeSafe's Jev vs. Stanford/NVIDIA's open CLM-8B, with CLM-8B beating Jev on several judge-pattern benchmarks (81.6% vs 71.1% DeepSWE, 87.6% vs 83.1% Terminal-Bench) while being free/self-hostable — a concrete new data point in RDCO's standing Jev due-diligence thread, not a paid Jev placement (disclosed sponsor was Ory, unrelated). Also covered independently the same day by Innermost Loop's issue.

AlphaSignal — 2026-09-28 new sponsor (added 2026-09-28)

• 2026-09-28-alphasignal-claude-physics-record-voicestudio-plugin-portal: The Linux Foundation is a new distinct sponsor (pool now 27+). Still unknown-until-scanned per issue.

AlphaSignal — 2026-09-29 new sponsor + masthead resolved as house self-promo (added 2026-09-29)

• 2026-09-29-alphasignal-sonnet55-terminal-bench-beats-opus5: AgentField AI (CodeAF harness) is a new distinct sponsor, standalone "Presented by" block (pool now 28+). Prior Labs and Google Cloud recurred (confirmed pool members). Separately, the masthead "In Partnership with" slot resolved this issue — but to AlphaSignal's OWN a16z Speedrun event promo (50% off tickets), not a third-party paid sponsor. Treat masthead resolutions as ambiguous between house self-promo and third-party sponsorship going forward — verify which before logging a masthead resolution as a new pool member.

AlphaSignal — 2026-09-30 new sponsor pair (added 2026-09-30)

• 2026-09-30-alphasignal-gpt61-sol-cost-cut-decisions-api: DigitalOcean & NVIDIA (joint promo, Open Intelligence Summit) is a new distinct sponsor pairing inside the Signals list (pool now 30+). WorkOS and Sentry both recurred (confirmed pool members). Masthead "In Partnership with" slot present but unresolved in plaintext, consistent with the majority-unresolved pattern.

AlphaSignal — 2026-10-01 new sponsor (added 2026-10-01)

• 2026-10-01-alphasignal-gemini4-argon-gpt61-sol-ultrafast: Multiverse Computing is a new distinct sponsor — native in-Signals-list self-promo ("Presented by Multiverse Computing") on its own Quasar 438B item. AgentField AI and Origin Technology both recurred (confirmed pool members). Pool now 31+. Masthead slot unresolved as usual.

AlphaSignal — 2026-10-02 two new sponsors (added 2026-10-02)

• 2026-10-02-alphasignal-claude-code-mods-codex-security-deepmind-consciousness.md: Fin (fireside-chat ad, distinct from existing pool member "Finest" — don't conflate the two) and Mercor (native in-Signals-list self-promo on the Qwen3.5 post-training item) are two new distinct sponsors (pool now 33+). DigitalOcean and NVIDIA both recurred (confirmed pool members). Masthead "In Partnership with" slot unresolved as usual.

Not Boring — Reducto resumed at WDoO #213 (added 2026-10-02)

• 2026-10-02-not-boring-wdoo-213.md: Reducto returned as sponsor after breaking its 3-issue streak at #212 (Jack & Jill). One resumption isn't evidence of a restored standing relationship by itself — continue treating WDoO sponsors as unknown-until-scanned per issue. Also newly observed: a quarterly Runway video partnership (used for a curation item) plus an inline plug for Not Boring's own paid "Deep Dive on Airbound" piece — both self-referential cross-promo, distinct from the Reducto paid block.

Mostly Metrics — 2026-10-01 new sponsor (added 2026-10-01)

• 2026-10-01-mostlymetrics-quote-to-cash-part-1 (Part 1 of a new 3-part Quote-to-Cash series): Tabs (billing-automation) is a new third-party paid sponsor — masthead "proudly powered by" block plus a UTM-tagged billing-lag-calculator CTA, topically adjacent to the issue's own billing-lag subject matter. Not previously in the known pool (Brex, Intuit, Samsara, Rivian, MLB, Abacum, Mostly Talent, Koyfin, Rillet). Mostly Talent (CJ's own recruiting arm) also appeared per usual as self-promo, not third-party. Watch Parts 2-3 for whether Tabs recurs as a series-standing sponsor (same pattern class as CFO Secrets/Stuut and CFO Secrets/Summation).