Specific
A cloud platform designed for AI coding agents — instead of a human clicking around a dashboard, your AI assistant can provision databases, storage, and APIs itself.
🔗 Visit Specific
Description
Tools like Cursor or Claude Code are good at writing application code, but building the backend — the database, file storage, authentication, deployment — usually still means a human switching over to a cloud console and clicking through setup screens. Specific removes that handoff: it's a cloud platform built specifically so an AI coding agent can create and manage the backend infrastructure itself, from a database to file storage to a live deployment, through natural-language instructions.
Specific offers a full backend stack — PostgreSQL with point-in-time recovery, S3-compatible object storage, Redis, vector databases, autoscaling APIs, CDN/TLS-fronted frontends, and Temporal-based workflow orchestration — accessible via an MCP server that any coding agent can connect to, or via specific dev (local environment) and specific deploy (production) commands. Pricing starts free (5 projects, 512MB DB, 1GB storage) and scales to $25/month (Pro) or $299/month (Scale) plus usage; infrastructure runs on AWS via Neon (Postgres) and Tigris (object storage). It's a Y Combinator F25 company founded by ex-Stripe engineers.
💬 Our review
The short version: Specific is chasing a genuinely new niche — infrastructure that's designed to be operated by an AI agent rather than a human — and if you're already leaning heavily on Claude Code or Cursor for full-stack work, it removes a real friction point.
Compared to Supabase or Vercel, which are excellent backends but still assume a human is configuring things through a dashboard, Specific's MCP-server-first design is built around agent workflows from the ground up. The tradeoff is a less mature ecosystem and smaller community than Supabase, plus a usage-based pricing structure that's harder to predict than a flat monthly fee — the free tier's 512MB database and 5-project cap will feel tight fast for anything beyond a prototype. For solo developers or small teams building MVPs primarily through AI coding agents, it's worth trying against the free tier before committing; teams that want a proven, human-first backend platform with a huge community are still better served by Supabase.
📊 Global score
🤖 AI-enriched data
Free : 5 projets, 512MB DB, 1GB storage. Pro : $25/mois + usage. Scale : $299/mois + usage. Enterprise : sur devis.
Pros
Conçu spécifiquement pour être piloté par un agent IA (serveur MCP natif)
Stack complète incluse : DB, storage, Redis, vecteurs, workflows
Déploiement edge avec CDN/TLS intégrés
Sauvegarde PostgreSQL avec point-in-time recovery
Cons
Tarification à l'usage difficile à prévoir
Plan gratuit limité (512MB DB, 5 projets)
Écosystème et communauté plus petits que Supabase