ViewDAO
Your scheduled posts are dying and you haven't noticed yet
Go to your X analytics right now. Pull up your last 30 scheduled posts. Look at which ones actually got engagement. Now compare that to the few times you tweeted something live, from your phone, about something that had just happened.
The live posts won. They always do. And you already knew that, you just kept scheduling anyway because that's what the tools are built for.
Buffer, Hootsuite, Later, Hypefury, Postiz. Every one of them solves the same problem: post consistently. Drag posts into time slots on a calendar grid. "Set it and forget it."
That was the right approach before recommendation engines took over the timeline. It stopped being right around 2023. A scheduled post is a bet that the conversation you care about will still be the conversation in six hours. On X in 2026, that bet loses more often than it wins. The ETF flow story that was hot at 9am is stale by 2pm. Your scheduled take on Bitcoin's breakout goes live forty minutes after Bitcoin already broke out, and it lands in a feed where three KOLs already said it better and moved on.
The creators actually growing on X right now are not posting more. They're posting faster. That's the whole game.
I'm going to walk you through how we built ViewDAO Studio's autopilot to solve this, and how to set it up so your account is posting about breaking news while you sleep.
STEP 1 — Stop treating all your content the same
You have two types of content and you're shoving both of them through the same scheduling funnel.
Evergreen content is the stuff that doesn't expire. A deep-dive on stablecoin regulation. A framework for evaluating L2s. A thread about how you think about portfolio construction. This content's half-life is weeks. People find it through search, citations, reshares. It doesn't matter if you publish it at 10pm Tuesday or 10am Wednesday.
Reactive content is the stuff that expires in minutes. Your take on the SEC chair's surprise statement. Your reply to the KOL who just dropped a hot take. Your read on the merger announcement that just leaked. This content is worth everything in the first hour and almost nothing after that.
Scheduling works for evergreen. It does not work for reactive. You need two separate lanes and you need to stop pretending one tool handles both.
ViewDAO Studio runs both lanes. The schedule handles your evergreen. Autopilot handles your reactive. That split is the whole product.
STEP 2 — Set up your voice profile before anything else
Autopilot can only sound like you if it knows how you sound. This is not optional.
Go to Voice in ViewDAO Studio and do one of two things:
Option A: Pick a preset. ViewDAO Studio ships with pre-built voice profiles for common creator types in crypto and AI. If you're just getting started and want to test autopilot quickly, pick one that's close to your style. You can swap it later.
Option B: Extract your voice. Paste 10 of your own tweets. Not random ones. Pick the 10 you're most proud of from the last month or two. The voice analyzer runs them through Claude Sonnet and pulls out a structured profile: tone scores, signature phrases, avoid phrases, sentence style, structural patterns. This gets injected into every post autopilot generates.
Option B is better. The posts it produces will sound like you specifically, not like a generic version of your niche. We covered the full extraction process in our Brand Voice vs. System Prompts article if you want the details on how this works.
Pick your 10 best recent posts. Run the extraction. Move on.
STEP 3 — Turn on autopilot and configure it
Go to Settings, then Autopilot. Here's what to set:
Flip breaking_news_enabled on.
Set breaking_news_max_per_day to 2. This is the daily cap. The system will not post more than 2 autopilot posts in a day no matter how much news breaks. Start here. You can raise it later if you want.
Set breaking_news_min_score to 6. This is the slop gate. Every post autopilot generates gets scored 0-10 across 7 dimensions before it can publish. Anything above 6 gets held as a draft for you to review instead of posting live. If you want to be stricter, set it to 7 or 8. At 8, most breaking stories won't produce a post that clears the bar. That's fine. The point isn't volume. The point is being early on the stories that are actually yours.
Set your niche. Crypto, AI, fintech, whatever you cover. This tells the trend intelligence which scopes to monitor for you.
That's it. Autopilot is live.
STEP 4 — Use the schedule for everything that's not time-sensitive
ViewDAO Studio has four daily schedule slots by default: 9am, 1pm, 6pm, 10pm Eastern. Use these for your evergreen content.
Long-form articles. Frameworks. Explainers. Reddit posts that need to hit a specific subreddit's peak hour. Threads you've been working on all week. All of this goes in the schedule.
Do not try to schedule reactive takes. Do not write a post about a news event and schedule it for tomorrow at 2pm. By 2pm the conversation will have moved. That's autopilot's job.
The schedule is your consistency engine. Autopilot is your relevance engine. They run in parallel.
STEP 5 — Understand what's happening under the hood
This is the part for the vibe coders who want to see the pipeline. If you just want to use the product and not think about how it works, skip to the next step.
Every 30 minutes, a Supabase cron job wakes up and runs this sequence:
At :00 and :30 past the hour, trend-cache-refresh hits xAI Grok with an x_search tool call across 8 scopes (crypto, AI, macro, AI stocks, DeFi, geopolitics, regulation, startups) and writes hot topics to stance_trend_cache. Each topic gets a heat label: hot, rising, or steady.
Five minutes later at :05 and :35, news-feed-refresh reads those hot topics, extracts 3-4 keywords per topic, and queries NewsData.io, CryptoPanic, and YouTube for fresh articles matching them. Any article with a reply_score of 8 or higher gets flagged breaking = true with a timestamp.
At the next :00 or :30, the autopilot function wakes up in breaking_check mode. It finds breaking articles in a 6-hour window, matches article titles against hot topic keywords (requires 2+ word overlap to prevent spurious matches), checks the daily cap, checks the 3-hour gap rule so you're not posting twice about the same event, fetches the full article body via Jina Reader, and generates a post with your brand voice applied and the full anti-slop filter run against it. If the slop score passes your threshold, the post ships to X immediately. If it fails, it goes to a review queue.
End to end from news breaking to post live: around 35 minutes on the outside, often closer to 15 when the news lands mid-cycle.
That's a different product from a scheduling tool. Scheduling asks "when do I want this to post?" This pipeline asks "what should I be posting about right now that I couldn't have known about six hours ago?"
STEP 6 — Know what the slop gate actually catches
Every autopilot post runs through slop-scorer, a Claude Haiku pass that scores on 7 dimensions: original take, no AI tells, voice match, platform fit, keyword density, no engagement bait, no emotional inflation.
The scorer has seen the full anti-slop rulebook. 60+ banned phrases. The em dash rule. The colon overuse rule. The "it's not just X, it's Y" structural tells. The inflated-adjective list: pivotal, robust, groundbreaking, nuanced, multifaceted, and the rest. If any of those appear, the score drops and the post does not ship.
The strongest version of the counterargument against autopilot is this: an AI posting in your voice about topics you didn't pick will eventually say something you wouldn't have said, and one bad post can damage your account permanently. That fear is real. That's why the slop gate exists. The system is designed to be conservative. It would rather miss a trend than post badly about one. Default max is 2 posts per day, and every one of them has to pass the scorer before it goes live.
STEP 7 — Let it run and watch what happens
Once autopilot is on, here's what your first week should look like:
Your scheduled posts go out at the times you set. Consistent presence, same as any scheduling tool.
When breaking news hits in your niche, autopilot fires. A post generates in your voice, gets scored, and if it passes, it publishes while you're asleep or at your desk doing something else. You wake up to a post that already has replies because it went out when the story was still fresh.
That's the part that scheduling can't do. On X, engagement on any trending topic concentrates in the first wave of accounts that post about it. The accounts that get their take out early collect the conversation. The accounts that show up an hour later land in a timeline where the discussion already moved past them. Autopilot gets you into that early window without you being online.
The uncomfortable part
If you are a creator on X and you are still running a scheduling-only workflow in 2026, you are competing against people who are using breaking news autopilots. Your ideas might be better. Doesn't matter. Theirs are live on the timeline while yours are still in a queue waiting for 1pm Tuesday.
Scheduling was the right tool for the era before recommendation engines took over. The queue is a lie in the way that all inherited defaults are lies: it was designed for a world that no longer exists, and most people haven't noticed because nobody told them.
The tools are cheap enough now that you don't have to choose. Run both lanes. Evergreen in the schedule. Reactive in autopilot. Stop treating them like the same product.
Try it at https://viewft.com/ and drop your results below. I'll answer every question.