The algorithm doesn't care about your posting schedule.
We decoded the open-source X algorithm. Everything you've been told about growth is wrong. One reply chain = 150 likes worth of reach.
Join the waitlist β’ No spam, ever
// scored_tweets_model_weight β published April 5, 2023
Bookmark is not in this formula. 19 signals are scored in weighted_scorer.rs β bookmark is not one of them.
A creator with 50,000 followers shared this with us last week.
"I've been posting 3x daily for 6 months. Engagement dropped 40%. I tried posting at 'optimal times,' using trending hashtags, even paid for courses. Nothing worked. My reach just... died.
Then I found out I was shadowbanned for 3 weeks and had no idea. No notification. No warning. Just silence."
Average engagement drop on X in 2025
Median engagement rate (3 per 10K followers)
Of full-time creators struggle to survive
We built the first AI that actually understands the X algorithmβbecause we read the source code.
See exactly how the algorithm will rank your content before you post. Real-time scoring based on actual algorithm weights.
Know instantly if you're being throttled. Get specific diagnosis and a recovery planβnot just "wait it out."
The highest-weighted signal in the algorithm: when your reply gets engaged by the author. We show you which replies to prioritize.
X open-sourced their algorithm. We did what no one else bothered to do: actually read it.
X's recommendation system uses a Grok-based transformer that predicts engagement probabilities. We mapped every weight, every multiplier, every penalty.
Our AI simulates how the algorithm will score your content before you post. No more guessingβsee your predicted reach in real-time.
Every prediction is tested against actual X performance. Our model learns and improves with every post analyzed.
// Final Score = Ξ£ (weight_i Γ P(action_i)) β from xai-org/x-algorithm
// params.rs excluded from open-source release β numeric weights for these 9 signals are not public
vqv_score only applies if video_duration_ms > MIN_VIDEO_DURATION_MS. Videos below this threshold receive zero VQV weight. β weighted_scorer.rs:vqv_weight_eligibility()
Bookmarks do not appear in weighted_scorer.rs. The function scores exactly 19 signals β bookmark is not one of them.
Per RETRIEVAL_SIGNALS.md, bookmarks are used as a feature in only 1 of 6 candidate sourcing systems (USS). By comparison: likes are used in all 6, retweets in 5 of 6, replies in 5 of 6.
Waitlist members get 50% off for life. No commitment until we launch.
For individual creators
For serious creators
For teams & agencies
Join the waitlist for early access and 50% off for life.