The problem
He tried Taplio and its alternatives. He says, “None of them fit how I actually research, draft, and iterate.” He went manual instead. He spent 45 to 90 minutes a day scrolling LinkedIn and X, bookmarking posts and pattern-matching by hand. In his words, “When you’re publishing 5 times a week, that math gets ugly fast.”
He says the bigger problem was the method. “Staring at a feed hoping for inspiration is a terrible ideation process.” He wanted something that surfaced patterns from real engagement data instead of whatever the feed showed him that morning.
The blueprint (as he describes it)
- Build it locally. He describes eight hours of focused work: four for the MVP and four more for polish across a week. All seven features run locally on his machine on a localhost server. They talk to LinkedIn through Apify’s API.
- Morning briefing. The first screen pulls his engagement stats from the last 24 hours and the trailing week. It surfaces his top three performing posts and flags content pillars he hasn’t covered recently.
- Research scraper. It finds the top-performing AI and marketing posts on LinkedIn from the past 24 to 72 hours and sorts them by real engagement. He says he would “rather see a 200-comment thread from a peer I respect than a 50,000-impression post from a guru clipping someone else’s idea.”
- Trend tracker. It clusters the topics getting the most engagement in his niche and surfaces themes picking up momentum. It flagged “job market disruption and the junior marketing crisis,” a topic he says was not on his radar.
- Lookalike content generator. It takes his top-performing posts and surfaces adjacent angles instead of rewrites. For a post about hiring globally, it might offer five: the operations, cost, talent quality, timezone and cultural sides. He picks the one he has a real opinion on and writes it himself.
- Narrative arcs. He calls this “probably the most useful feature in the whole tool.” It shows how long it has been since he posted about each pillar. His example dashboard shows Claude Marketers Academy at 8 days and GrowthPair at 33 days. He rotates between 3 to 5 pillars a week based on that data.
- Content recycler. It identifies past top performers worth reposting. He ties this to narrative arcs to time reposts.
- Engagement scraper. It finds people engaging with content in GrowthPair’s ICP: “VPs of marketing, founders, growth leads at venture-backed companies.” He gets a list of warm signals every morning that he can DM. He calls this feature “still a bit experimental.”
- Write and post by hand. He prototyped a “draft the post for me” feature and killed it inside a week. He writes: “Content Machine 2000 doesn’t write a single word that goes on my LinkedIn.” He also skipped auto-posting and auto-scheduling because programmatic posting “carries real risk of LinkedIn flagging the account.” Manual posting takes him 30 seconds. He left out notifications on purpose.
His typical Monday, as he lays it out:
- 6:30 AM: he opens the dashboard. The briefing shows yesterday’s post at 4,200 impressions and his GrowthPair pillar at 33 days (overdue).
- 6:40 AM: he pulls an angle from the lookalike generator, writes a draft and publishes it on LinkedIn.
- 7:00 AM: the post goes live and he responds to comments for 20 minutes.
- 9:00 AM: he checks the engagement scraper. Three names from real target companies showed up, and he sends two short, specific DMs about their comments.
His build advice: document what you do every Monday morning and ask Claude to build it. Tell it to “ask me clarifying questions before you start writing code.” Feed it your past posts, analytics exports, style guide and ICP definitions. Build the MVP in one afternoon and live with it for a week before adding features. Block out 2 to 3 hours for the first session.
Stated results
- A repost about Claude Marketers, timed with narrative arcs, hit 101,730 impressions.
- A post about junior marketers, prompted by the trend tracker, beat his typical impression count. He usually lands 2,500 to 3,500 per post. The exact figure for that post is not stated.
- Reposts typically pull about 80% of the original engagement on his account.
- Content time is about 20 minutes a day. Before the tool, “it was closer to 3 hours, most of it scrolling.”
- The gap between deciding what to post and having the topic “closed from 45 minutes to about 90 seconds.”
- It replaced four SaaS subscriptions. The stack costs about $15 a month across Apify API calls and Claude. The tools it replaced ran roughly $50 to $150 a month each.
- He says it added “much more pipeline for GrowthPair” but calls this hard to quantify. No pipeline number is stated.
Caveats
- All figures are self-reported from his own account. The 101,730 repost and the 4,200 Monday post are single examples, not averages.
- GrowthPair appears throughout as a pillar and as the ICP for his outreach. Per the verified notes, it is his company. The pipeline claim is unquantified.
- He says he built it “in a weekend,” but also describes polish “across a week.” The total is stated as eight hours.
- The post ends with a pitch to build your own tool. It does not sell a product, course or kit. The Substack page claims “Over 18,000 subscribers” in its site chrome; that is the publication’s figure, not a result of this system.
- GStack is Garry Tan’s Claude skills, mentioned as a recommendation. No partnership is stated with Apify, Anthropic or GStack.
- This is a custom research dashboard built with Claude Code. It runs scrapers and surfaces data. It is not an autonomous agent that writes or posts. The Claude cost is described only as “a small Claude subscription overhead.”
- Handles: X @jon4growth, Bluesky jon4growth.bsky.social, LinkedIn linkedin.com/in/jon4growth, Substack jon4growth.substack.com (all verified 2026-11-05). Threads, Instagram, TikTok, Mastodon: none verified.
FAQ
Does Content Machine 2000 write his LinkedIn posts?
No. He prototyped a “draft the post for me” feature and killed it inside a week. He picks the angles and writes every post himself.
Does it post or schedule automatically?
No. He says programmatic posting carries real risk of LinkedIn flagging the account, and manual posting takes him 30 seconds.
What does it cost to run?
About $15 a month across Apify API calls and a small Claude subscription overhead. He says the tools it replaced cost roughly $50 to $150 a month each.
How long did it take to build?
He describes eight hours of focused work, four for the MVP and four for polish. He says he had zero engineering background and built it with Claude Code.
Sources
- https://jon4growth.substack.com/p/inside-the-linkedin-content-system (primary: first-person build write-up on his Substack, The AI Marketer, dated Jun 09, 2026)
Want agents like these working your distribution? Get It Built. Get new teardowns weekly: subscribe (double opt-in).
Get one teardown a week.
Same format, a new operator every issue. Double opt-in.