Why AI agents forget everything (and how we fixed it)

Published: (March 14, 2026 at 08:27 PM EDT)
2 min read
Source: Dev.to

Source: Dev.to

If you’ve built an AI agent or assistant, you’ve hit this wall: the moment the session ends, it forgets everything. The user comes back the next day, and the agent has no idea who they are—no memory of preferences, history, or ongoing work. The user has to re‑explain everything from scratch every single time.

This isn’t a model problem; it’s an infrastructure problem. Models don’t have long‑term memory—they only have context windows, and when the window closes, everything in it disappears.

What we built

AmPN is a hosted memory store for AI agents. Your agent stores memories via our API, and when a new session starts it retrieves the relevant context with semantic search—so it picks up exactly where it left off.

from ampn import MemoryClient

client = MemoryClient(api_key='your_key')

# Store a memory
client.store(
    user_id='alice',
    content='User prefers concise explanations and works in Python'
)

# Retrieve relevant context
results = client.search(
    user_id='alice',
    query='what does the user prefer?'
)
# Returns: ['User prefers concise explanations and works in Python']

The tech

  • Storage: PostgreSQL for structured data, Qdrant for vector embeddings
  • Search: Sentence embeddings (all‑MiniLM‑L6‑v2) — finds semantically similar memories, not just keyword matches
  • API: FastAPI, fully documented
  • SDKs: Python (pip install ampn-memory) and Node.js (npm install @ampn/memory)

Free tier

The free tier is open now at ampnup.com. It includes 1,000 memory operations and 10,000 reads per month—enough to build and test.

We’d love to hear how you’re handling memory in your agent projects today.

0 views
Back to Blog

Related posts

Read more »

Travigo

Travel as fast as you speak with Gemini! Where live agents meet immersive storytelling & 3D navigation. This project was created for entering the Gemini Live Ag...

Micro games

Hey Gamers! 👾 As part of the Rapid Games Prototyping module, we are tasked with reviewing a peer's game. The challenge is to analyse a prototype built in just...