The latest articles on DEV Community by Paul N (@paulfromgidstats). https://dev.to/paulfromgidstats https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3882646%2Fd383deec-30d3-4947-81a2-be01e0b41c8a.png https://dev.to/paulfromgidstats en Paul N Thu, 16 Apr 2026 14:29:01 +0000 https://dev.to/paulfromgidstats/i-built-an-ai-news-desk-for-my-mma-site-heres-what-actually-worked-5d94 https://dev.to/paulfromgidstats/i-built-an-ai-news-desk-for-my-mma-site-heres-what-actually-worked-5d94 <p>My first prompt was: "You are a sports journalist. Write a 400-word article about this UFC event." A friend texted me "bro this reads like a bot wrote it." Three published pieces in. Fair.</p> <p><strong>The Stack</strong></p> <p>GidStats.com runs on a custom CMS with structured fight data across UFC, PFL, LFA, DWCS. The news pipeline sits on top of that. Playwright scrapes official promotion feeds, a webhook fires a Claude API call with relevant database stats as context, output lands in staging for review before anything goes live.</p> <p><strong>The Prompt</strong></p> <p>You need to specify what to avoid as precisely as what to produce. Mine has a 75-term banned-words list - "delve," "nuance," "tapestry" and everything else that screams AI. Explicit sentence structure instructions. Style references: ESPN MMA, The Athletic, Bloody Elbow.</p> <p>Structured fighter data gets injected before the generation instruction. Records, finish rates, physical attributes. The model can't hallucinate stats it's already been given, and MMA fans will fact-check you.</p> <p><strong>Review</strong></p> <p>Early on I reviewed for accuracy. Wrong question. The right question is "would I actually read this <a href="https://gidstats.com/news/" rel="noopener noreferrer">MMA news</a>?" I rebuilt the checklist around that: does the first sentence pull you into the second, does the fight analysis say something tactical rather than obvious.</p> <p><strong>What It Can't Do</strong></p> <p>Breaking news at speed - the pipeline isn't a 10-minute turnaround. Anything emotionally weighted: injuries, retirements. The model produces something flat where a human voice is the whole point.</p> <p>What I'd Do Differently</p> <p>Start with the voice problem, not the infrastructure. I spent two months on scraper architecture before solving what the output should sound like. Get one piece reading exactly right, reverse-engineer the prompt, then build the pipes. Also: inject structured data as context, always. Model memory is wrong often enough to matter.</p> ai beginners news