Case study
GetF4F — TikTok Social Growth Exchange
Organic Creator Growth. Algorithm-Safe Reciprocal Matching.
Full-Stack Architect & Lead Developer — P2P Social Growth Exchange & Proof Verification Engine.
A scalable, organic peer-to-peer social growth platform that enables TikTok creators to safely exchange followers, likes, and engagement through heuristic proof analysis, asynchronous reciprocal matching, and a self-healing credit economy.
Overview
GetF4F is a production-grade social growth platform designed to help TikTok creators overcome the critical 1,000-follower threshold needed for TikTok LIVE and monetization features. Operating without an official TikTok follow-verification API, GetF4F uses an asynchronous delayed-matching queue, heuristic screenshot validation, and a self-healing Trust Score economy to protect creators from fake bots and algorithmic shadowbans.
The problem
TikTok creators struggle to unlock LIVE features, buying fake bots results in shadowbans, and mutual following within seconds triggers TikTok's automated anti-spam flags. Furthermore, TikTok provides no commercial API to verify follower relationships.
The solution
A rule-governed peer exchange platform: tasks are assigned asynchronously with a 24-hour buffer, proof screenshots are verified with heuristic checks, and users maintain a 100-point Trust Score with automated penalties for unfollowers.
- Asynchronous Delayed Matching: Matches creators 24 hours apart to mimic organic discovery and eliminate reciprocal follow flag detection.
- AI-Assisted Screenshot Verification: Validates MIME payloads, minimum dimensions, profile button states ('Following' vs 'Follow'), and handles.
- Self-Healing Trust Score: All users start at 100 Trust Score. Freeloaders or unfollowers lose 15 points; accounts below 60 are automatically banned.
- Double-Entry Credit Ledger: Every credit earned, spent, or refunded is tracked with immutable transactional integrity.
- Admin Operations Suite: Global emergency kill switch, live cooldown settings manager, spot-check audit queue, and administrative activity logs.
Technical architecture
Asynchronous Queue & Anti-Bot Rate Limiting
Server actions process task matching with cooldown windows and historical pair checking. Prevents duplicate pairings and enforces dynamic velocity limits (5-20 matches/day) to keep creators within safe social algorithmic parameters.
Hybrid Consensus Verification Pipeline
Combines immediate client-side screenshot analysis with a 5-day grace period. If a user unfollows during the grace period, peer reporting triggers evidence review before credit balances finalize.
Relational Schema with Drizzle ORM on Neon Postgres
Utilized Drizzle ORM's type-safe query builder on Neon Serverless PostgreSQL for transactional credit mutations, user status state machines, and relational match logs.
Progressive Web App & Multi-Channel Alerts
Configured installable PWA manifest with background service workers, transactional emails via Resend, and automated WhatsApp match notifications.
Problems I ran into
…and how I worked through them.
The problem
Closed Social Ecosystem: TikTok offers no commercial API to programmatically verify whether User A followed User B.
How I solved it
Engineered a hybrid consensus model: client screenshot heuristic analysis paired with a 5-day grace period and peer accountability.
The problem
Algorithmic Reciprocal-Follow Penalties: Rapid instant follow-backs trigger TikTok anti-spam filters.
How I solved it
Built a time-delayed queue engine that defers follow-backs by 24–48 hours, blending actions into natural creator discovery patterns.
The problem
Freeloader Exploits: Bad actors attempting to gain followers without contributing follows to the community.
How I solved it
Enforced strict credit-gated queue access requiring earned credits before entering the follower queue, backed by -15 trust penalties.
Results
- Architected full-stack application with Next.js 16 Server Actions, Neon PostgreSQL, and Drizzle ORM.
- Protected creator accounts with 24-hour delayed matching algorithms mimicking natural human activity.
- Cut fraudulent task submissions and manual moderation overhead via heuristic screenshot validation.
- Shipped complete administrative operations dashboard with live kill switches and audit trails.