Your AI Developer Agent
Neo is a persistent AI developer that remembers context, understands codebases, and follows structured development methodology. Deploy your own instance in minutes.
Remembers context across sessions. No more re-explaining your project every time.
Understands your code structure, dependencies, and hotspots before making changes.
Follows Superpowers 7-phase workflow. Tests first, always.
Neo uses a 5-layer memory architecture that never loses context — even across restarts, compactions, and long gaps between sessions.
flowchart TB
subgraph ELITE["🧠 Elite Longterm Memory"]
direction LR
HOT["🔥 Hot RAM
SESSION-STATE.md
survives compaction"]
WARM["🌡️ Warm Store
LanceDB vectors
semantic search"]
COLD["🧊 Cold Store
Git-notes graph
permanent decisions"]
end
MEM["📖 MEMORY.md + daily/
curated · human-readable"]
HOT --> MEM
WARM --> MEM
COLD --> MEM
SESSION-STATE.md survives context compaction. Write-Ahead Log protocol ensures nothing is lost mid-conversation.
LanceDB vectors enable semantic search. "Find everything about auth" works even without perfect categorization.
Git-notes knowledge graph for permanent decisions. Tamper-evident, auditable, travels with your repo.
Token-efficient memory that grows with your project. Automatic summarization keeps context windows lean.
Raw daily notes get distilled into MEMORY.md. Recent detail, older summaries.
Weekly reviews merge learnings. Token cost stays flat as project grows.
Full-stack, TDD, code review, refactoring
Clinical trials, FDA regulatory, medical devices
Dealer operations, inventory, payments
Browse every guide: setup, architecture, memory, domains, and launch operations.
How to create a Personal Access Token
How Neo fleet works
Multi-tenant agent deployment
Specialized agent configurations
5-layer memory architecture deep dive
Token-efficient indexing strategy
Built by The Menon Lab