ANNEX DOC. AIL-301 | PUBLIC SYSTEM DOSSIERS
Agentic Intelligence Lab Persistent Agent Systems
Volume Ecosystem
Systems 2 Active
Mode Dossier
SYS: PERSONAL_MODEL_FIELD X:--- Y:---
FIG. E1 Continuity and correction as a vector field.
SYS: ROUTING_MATRIX X:--- Y:---
FIG. E2 Routing decisions under explicit constraints.

Systems in the Loop

These projects are not showcases.

They are test benches for agent memory, routing, inference, and deployment.

Flagship Project

Elephant Agent

A personal-model-first agent for durable, correctable memory.

  • Correctable user model instead of a hidden profile.
  • Curiosity that asks only when the answer changes future help.
  • Continuity across CLI, chat, dashboard, messages, tools, and background learning.
Website GitHub
Type Personal-model-first AI agent
Scope Personal Model, curiosity, memory, and continuity
Focus Correctable understanding, local-first operation, and user-paced learning

Project

vLLM Semantic Router

An open routing layer for choosing the right model, tool, and execution path.

  • Routes across heterogeneous model pools.
  • Trades cost, latency, privacy, safety, and quality explicitly.
  • Extends inference routing toward agent orchestration.
Website GitHub
Type Open-source routing system
Scope Inference, model selection, and agent-facing control flows
Focus Cost, latency, privacy, safety, and execution quality

Research Fit

Find the sharp systems question.

Start where agent systems break.

Build Surface

Ship something public.

Interfaces beat internal demos.

Deployment Loop

Let workloads judge it.

Failure is input to the next paper.