# Project export: Donala

This document was generated by HackStack to give an AI agent context about a hackathon project. Sections are labeled with their provenance; content marked as truncated was cut to keep this document small.

## Project metadata

- Hackathon: OpenAI Build Week
- Tagline: Persistent memory infrastructure for AI agents that keeps context, powers recall, and helps apps remember users across sessions. That is Donala, at your service
- Devpost: https://devpost.com/software/donala
- GitHub: not linked
- Demo: https://donala.guardianity.space/
- Video: https://www.youtube.com/embed/3xKVw5GPOaU?enablejsapi=1&hl=en_US&rel=0&start=&version=3&wmode=transparent
- Team: contributor stats unavailable

## Devpost submission (written by the team)

### Inspiration

AI agents lose context between conversations. Donala was built to give them secure, durable memory without forcing developers to manage raw chat histories in every prompt.

### What it does

Donala provides persistent memory for AI agents through simple write and recall APIs. It supports: UniMemory for shared organizational knowledge MultiUser Memory for safely partitioned, per-user context PostgreSQL-backed production storage Memory instructions, scheduling, locks, audit activity, and usage controls Nexo, an AI assistant for marketing workflows ##

### How we built it

We built the product with a Next.js dashboard, Django/DRF API, PostgreSQL integrations, and a high-performance Rust memory runtime. Memory records are encrypted, scoped to a memory space, backed up to the client-connected database, and retrieved through a query-driven recall workflow.

### Challenges we ran into

The hardest parts were making memory durable without sacrificing speed, separating user data correctly in multi-user applications, securing database credentials and API access, and designing a dashboard that makes infrastructure setup approachable.

### Accomplishments we're proud of

We created an end-to-end memory platform: developers can create a memory space, connect a database, issue an API key, write conversation turns, and recall relevant context in their agent loop. We are especially proud of the clear separation between shared and per-user memory, encrypted records, owner-controlled AI permissions, and production backup support.

### What we learned

Persistent memory is more than saving chat logs. Useful memory needs reliable storage, relevant retrieval, privacy boundaries, durable identifiers, clear ownership, and safeguards so recalled data is treated as reference material rather than model instructions. ##

### What's next

Next, we want to expand database support, improve retrieval quality and observability, add more agent-framework integrations, deepen Nexo’s automation capabilities, and make onboarding even faster for teams deploying production AI agents.

## README (from the GitHub repository)

No README available.

## Detected evidence (automated analysis)

No repository was indexed for this project. Claimed technologies below could not be checked against code.
- Python (language) — claimed on Devpost, not found in the code
- Rust (language) — claimed on Devpost, not found in the code

## Codebase structure

No repository index available.

## Key source files

No repository index available; no source files included.