Preference learning for AI agents
Your agent should
learn preferences
Users correct their agents every day, then the session ends and it's forgotten. pref0 makes those lessons stick.
First 100 requests/mo free
Users correct their agents every day, then the session ends and it's forgotten. pref0 makes those lessons stick.
First 100 requests/mo free
Same preference, different sessions. Confidence compounds until the agent just knows.
Real signals pref0 extracts and compounds across conversations.
"Use TypeScript, not JavaScript"
language: typescript0.70"Deploy to Vercel, not Netlify"
deploy_target: vercel0.70"Use pnpm instead of npm"
package_manager: pnpm0.70"Bullet points, not paragraphs"
response_format: bullet_points0.70"Keep it under 5 lines"
response_length: concise0.40"Use Postgres, not MySQL"
database: postgres0.70Each preference starts with a confidence score. Repeat it across different conversations and it becomes a strong learned preference.
Three API calls. The learning happens automatically.
Pass chat history after each session. pref0 extracts corrections and preferences automatically.
Same preference across sessions? Confidence goes up. The profile gets sharper over time.
Fetch learned preferences before your agent responds. It behaves like it already knows the user.
1. Track a conversation
await fetch("https://api.pref0.com/v1/track", {
method: "POST",
headers: {
Authorization: "Bearer pref0_sk_...",
"Content-Type": "application/json",
},
body: JSON.stringify({
userId: "user_abc",
messages: conversation.messages,
}),
});2. Fetch preferences at inference
const res = await fetch(
"https://api.pref0.com/v1/profiles/user_abc",
{ headers: { Authorization: "Bearer pref0_sk_..." } }
);
const { prompt } = await res.json();
// → "Prefers TypeScript, pnpm, Tailwind..."requests every month
to reach high confidence
to integrate with any agent
Memory stores logs. RAG retrieves documents. pref0 extracts structured preferences from corrections, compounds confidence over time, and serves them at inference.
| Memory | RAG | pref0 | |
|---|---|---|---|
| Stores | Raw conversations | Documents | Structured preferences |
| Learns over time | No | No | Yes, confidence compounds |
| Handles corrections | No | No | Core signal |
| Integration | Varies | Vector DB + retriever | 2 endpoints |
| Scoping | Per user | Per collection | User → Team → Org |
Your users are already teaching your agent what they want. pref0 makes sure the lesson sticks.
100 requests/mo free · $0.005 per request after