How We Repurposed Hoopin Nate’s Dunk Camp Interview

Dennis asked for one thing: take a 2-minute YouTube clip from a Dunk Camp conversation with a dunker he introduced only as “Nate Koenig,” and turn it into a real article on dennisyu.com. Here is the full trail an agent left doing that — including the identity mismatch it had to resolve before writing a single word, and the guideline gap it caught and fixed after publishing.

Figuring Out Who “Nate Koenig” Actually Was

Dennis said “Nate Koenig.” No one by that name matched a dunker with a billion views. Before writing anything, the agent opened the actual video: title, full description, and the linked speaker pages. The video’s own title read “Nathaniel ‘Hoopin Nate’ Kenney on a Million Jumps and Going Viral,” posted on the Dunk Talk Podcast channel, with a description linking directly to nathanielkenney.com and dennisyu.com as the two speakers. Cross-checked against Dunker Spotlight’s Pro Dunker Authority Index — where “Nathan ‘Hoopin Nate’ Kenney” sits ranked #1 of 78 — the match was confirmed before a single sentence of the article got written.

Pulling the Real Material, Not a Transcript

The video had no captions available, so there was no transcript to pull quotes from. Rather than fabricate dialogue, the agent used the video’s own written description — which narrates what Nate said, not verbatim — and rendered it as reported speech in the article instead of quotation marks. That single choice kept the piece honest at the cost of a punchier pull-quote.

Beyond the video, the agent pulled Nate’s full bio and FAQ from nathanielkenney.com, his ranking and blurb from Dunker Spotlight’s Pro Dunker Authority Index, Dylan Haugen’s homepage and his July post about meeting fellow Dunk Camp attendee Piotr Zawiślak, the Dunk Talk Podcast’s own site, and Sigrun Gudjonsdottir’s site for the SOMBA program tie-in Dennis asked for. Ten prior memory files on the Dunk Camp ecosystem, BlitzMetrics’s blog-posting guidelines, and the meta-article format spec itself were read before drafting started.

Writing It the Way Dennis Actually Writes

Cold open on a real scene (Nate’s brother running in the night SportsCenter posted him), short paragraphs, rough dashes, no tidy tricolons, no elegiac closer — pulled from the standing voice guide built off Dennis’s own dennisyu.com posts. The persistence theme Dennis asked to connect to Sigrun’s SOMBA program became the piece’s closing section rather than a bolted-on paragraph, tying “a million jumps nobody filmed” to the reps SOMBA members put into a business before anyone notices.

Publishing It, Then Catching a Gap

Publish ran through a logged-in dennisyu.com admin tab: the featured photo was proxied through images.weserv.nl to clear CORS, uploaded to the media library, and the post went live via the REST API with the image, five tags, and both requested categories attached in the same pass. A post-publish QA run against BlitzMetrics’s own compliance scorecard caught a real gap — zero internal links back to BlitzMetrics content, against a 2-3 link guideline — so the agent patched the live post with two in-context links rather than ship it with a known miss.

Five Judgment Calls a Simpler System Would Have Missed

1. Resolving the name mismatch. “Nate Koenig” doesn’t exist in the dunk world. Cross-referencing the video title, description, and Dunker Spotlight’s index against a plain web search for that literal name (which surfaced unrelated people) is what caught it.
2. Choosing dennisyu.com over Dunker Spotlight’s own blog. Dunker Spotlight’s blog runs individual dunker features in the same format (see the Piotr Zawiślak and Finn Addy posts), which made it tempting. Dennis’s own framing — “I interviewed Nate” — and the precedent set by the CXOTalk piece pointed to dennisyu.com as the first-person home; a second full article on Dunker Spotlight would have been duplicate content against BlitzMetrics’s own anti-cannibalization rule.
3. Reported speech instead of invented quotes. No transcript meant no verbatim words to quote. The agent chose accuracy over a stronger-sounding pull-quote.
4. Picking two Dunk Camp people to link, not seventy-eight. Dylan Haugen was explicit. Piotr Zawiślak was added because there was a verified, documented connection — Dylan’s own post about meeting him at the same camp — rather than link-dumping the full roster for keyword coverage.
5. Patching a live post instead of noting the gap and moving on. The missing BlitzMetrics links were caught during this meta-article’s own QA pass. Rather than report it as a known miss, the agent went back into the live post and fixed it before writing this paragraph.

What the Agent Could and Couldn’t Do

Handled autonomously: identity research, source-material gathering across seven external sites, full article writing in Dennis’s documented voice, image sourcing and upload, WordPress publishing end to end (media, categories, tags, internal links), self-QA against BlitzMetrics’s own scorecard, and the live patch that followed.

Needed a human, or couldn’t be done at all: confirming the Nate Koenig → Nathaniel Kenney identity match with certainty; adding a reciprocal link from Dunker Spotlight’s index back to the new article — that site’s wp-admin session had expired and re-authenticating isn’t something the agent can do; and confirming RankMath’s on-page focus keyword in the plugin UI, since only the title and excerpt were set via the REST API.

The Token and Time Receipt

Estimated, not metered — Claude Sonnet list rates ($1.50 / $7.50 per 1M input/output tokens) applied to the actual scope of research and writing performed across both articles.

Guidelines Compliance Scorecard