Reported operating use · Creator publishing and newsletter operations
An author’s reported Substack notes planning and review workflow
The November demonstration reports a creator’s publishing work using his own product. Its July setup is historical support, not a new deployment. The proposed enterprise workspace below is separate and unexecuted.
Plan, review and schedule Substack notes from published articles
GenAI Unplugged (Dheeraj Sharma’s publication)
Author-reported publishing use and live demonstration, published 12 November 2026; a linked July setup predates this coverage.
A published third-party example, not a Nuvepro customer case or evidence of a partnership.
The workflow at a glance
Calendar-led note drafts receive review before scheduling
Illustrative summary of the published account
- Author and connector settings
Set the reported permission boundaries
The published demonstration describes automatic reads, with approval for deletion and subscriber export. These settings do not prove every write action is technically gated.
- NextAuthor and assistant
Inspect calendar capacity
The author asks how full the coming week is before deciding how many notes to prepare.
- NextAssistant and publication archive
Draft notes from a published post
The assistant uses an existing article to prepare note drafts for the author’s review.
- NextAuthor and scheduling connector
Review drafts, then schedule
The author reports a review stop before scheduling notes; the demonstration describes a note going live through the connector.
What the source reports
Dheeraj Sharma and Wyndo describe checking calendar capacity, drafting notes from published articles, reviewing drafts and then scheduling them. The author also demonstrates archive edits, analytics and inbox work. These are components of one publishing implementation, not nine deployments or a proven growth result.
Reported technology: SubflowAI Substack MCP, AI clients and author review. Scope and outcomes are source-reported.
What the enterprise environment would need
The following is a proposed project design for an enterprise doing similar work. Published details are marked; the remaining systems, data and scenarios would be agreed for your project.
Proposed practice environment
Prepare a checked publication calendar without sending
Illustrative project design to scope with your enterprise.
- 1 · Set up the environment
Fictional archive, voice rules and calendar
Use synthetic published posts, a frozen calendar/time-zone configuration, embargo rules, approved voice and claim references, and a versioned permission matrix. Keep live accounts and subscriber data disconnected.
- 2 · Preparation and agent contributions
AI drafts notes and a proposed slot plan
An agent links each draft to its source article and proposes unoccupied slots. It flags unsupported claims, repeated content and uncertain scheduling assumptions for the editor.
- 3 · Independent checks
Verify claims and calendar conflicts independently
A separate editor compares factual claims and voice with the approved source references. Independent calendar checks recompute date, time-zone, duplicate, embargo and slot-conflict rules against an editor-approved reference plan.
- 4 · People decide acceptance
The editor accepts the exact draft batch
The accountable editor reviews the independently checked text and calendar version, resolving voice and editorial questions before acceptance. Practice acceptance enables no live schedule, send, deletion or export.
A reviewable evidence package
A versioned draft-note batch and calendar proposal with article references, conflict checks and editor dispositions.
If a check fails → withhold acceptance.
A changed article, occupied slot, duplicate note, unsupported claim or editor rejection blocks acceptance. Repair the drafts or plan, repeat independent source and calendar checks, then return the revised batch to editor review.
Failure or owner rejection returns to review
- Hold
- → 2 · Revise work
- → 3 · Rerun independent checks
- → 4 · Owner review
Operating boundary. Designed checks, not executed. A disconnected draft-only practice environment. No live post, reply, archive edit, deletion or subscriber export; no growth or operating-control outcome claimed. The publication owner must confirm editorial and scheduling references.
- Fictional archive, voice rules and calendar snapshot
- Proposed: a publication owner provides approved synthetic articles, a voice reference, embargo rules and a versioned calendar with time zones and occupied slots. Define which source claims can be reused before an agent drafts.
- Draft notes and a proposed slot plan
- Proposed: an agent drafts source-linked notes and suggests available slots in a disposable calendar. Unknown claims and ambiguous times remain unresolved. No real connector, subscriber export or live publishing tool is enabled.
- Independent content and calendar checks
- Proposed: a separate editor checks claims and voice against approved source references; an independent calendar check detects duplicates, embargo conflicts, time-zone mistakes and occupied slots. A generated draft cannot verify itself.
- Editor acceptance of exact draft and slot versions
- Proposed: the publication owner accepts or rejects each independently checked note and its slot plan. Rejection or a stale calendar blocks acceptance; repair the content or plan, rerun independent checks and return the revised version for review. Acceptance remains draft-only.
A proposed simulation
An illustrative practice design derived from this work, not an executed assessment.
Proposed: draft source-grounded newsletter notes and a reviewable calendar plan from fictional articles, with independent editorial and slot checks.
Review gate: The publication owner accepts only the exact independently checked note and calendar-plan versions. Acceptance does not authorize scheduling, publication, replies, deletion or access to a real subscriber list.
Challenge the result: A calendar slot becomes occupied after checking. Withhold acceptance, repair the slot plan, recheck duplicate/embargo/time-zone rules and source fidelity independently, then return the new version to the editor.
What needs checking
- Source boundary: the creator promotes his own connector. Reported human review and permission settings are not independent proof of technically enforced controls or improved readership.
- Proposed: test a supported note, an invented source claim, a duplicate, an embargoed topic, an occupied slot and a daylight-saving ambiguity against a separately prepared editorial/calendar reference. Designed, not executed.
- Proposed: occupy a slot or revise an article after checking. Withdraw acceptance, repair the affected draft or slot plan, repeat independent content/calendar checks and return it to the editor. Designed, not executed.
Review environment fidelity, human capability and agent performance separately. The enterprise’s accountable owners decide acceptance of the shared workflow and any operating use.
How an enterprise could build similar work
A proposed build order for an enterprise doing similar work, written by Nuvepro from the published account and our interpretation of the environment and checks above. It is not how GenAI Unplugged (Dheeraj Sharma’s publication) built or operates this work.
- Agree archive, voice and slot referencesProposed: freeze synthetic articles, approved claims, voice guidance, embargoes and a time-zone-aware calendar snapshot.
- Draft notes and a slot planProposed: an agent produces source-linked drafts and suggested slots in an isolated workspace; leave missing evidence unresolved.
- Check editorial content and calendar independentlyProposed: separate checks verify factual support, voice, duplicates, occupied slots, embargoes and time-zone handling.
- Repair conflicts and repeat checksProposed: withhold acceptance after any failed check or changed input, repair content or slots and rerun independent verification.
- Editor accepts exact draft versionsProposed: the owner accepts or rejects the checked notes and plan for practice only. No real scheduling, publication or subscriber operation follows.
Your systems, data, rules and accountable owners decide the real project. Agree acceptance criteria, escalation owners and operating responsibility with the people accountable for the work before anything runs on live records. Treat the first run as a rehearsal on test records: compare the draft output with an independently prepared answer, and record what the people reviewing it change.
Questions about this example
- Do the nine demonstrated jobs mean nine deployments?
- No. They belong to one creator implementation. This example focuses on calendar-aware note drafts, human review and scheduling.
- Is the November article a newly built implementation?
- Not established. The author links to a setup account published in July 2026; recent coverage is distinct from implementation age.
- Are the source approval controls independently verified?
- No. The written accounts describe review practice and settings. This review did not inspect the connector or establish technical enforcement for every action.
- Can the proposed practice publish to a real account?
- No. It uses fictional articles and a disposable calendar, with independent editorial/slot checks and exact-version owner review. These checks are designed, not executed.
Shape this around your enterprise
Bring a similar workflow and the people responsible for it. Together, define the records and systems to use, the result your team will build or configure, and how its owners will check it. Nuvepro scopes the simulation, environment and preparation. Agree any existing-agent evaluation, implementation assistance and production responsibilities separately.
Sources and limitations
Curation review: . This is the date this source example was reviewed, separate from source publication dates.
- The public written accounts were read; the embedded demo video, private publication records and actual connector behavior were not inspected.
- The author promotes his product. Growth, saved effort and control effectiveness are not independently validated.
- The supporting setup is dated 14 July 2026 and describes an independent tool using an unofficial API. November coverage is not evidence of a newly built deployment.
- Notes receive reported human review; do not generalize delete/export approval statements into an audited approval rule for every action. Other demonstrated jobs are not a single sequence.
- The separate enterprise practice is fictional and unexecuted. It produces drafts only and grants no live scheduling, publishing, reply, deletion or subscriber-export authority.
- Automate Substack growth from Claude or ChatGPT: 9 live workflowsSource publication: 2026-11-12
- I Stopped Opening Substack. Now I Run the Whole Thing From Claude (or ChatGPT).Source publication: 2026-07-14
More sales and marketing examples
- Jakob Nielsen’s reported training-video production workflowJakob Nielsen (personal training-video production)
- Reported student-recruitment CRM work and unresolved handoffsPima Community College (as reported by Three Sonorans)
- AI ad campaign management: Advolve exampleAdvolve
- AI car buyer qualification: Cars24 exampleCars24