Roadmap to HAICOSYSTEM v0.0.1

[XuhuiZhou]

Description

In this version, our goal is to work on the pipeline to support the interactions among human agents; AI agents; LLM LLM-simulated environments. For v0.0.1, our goal is to support large-scale simulation and evaluation so that we can obtain some preliminary results for our project.

To achieve that, we need to work on the following components:

Stress testing the system:

• [FEAT]: stress test the system #6 Everyone should try the system following the instructions and report potential issues during scaled simulation

Fix existing bugs:

• Include a test for the revising function in grounded engine #2
• Ensure agent action is correct #1

Please add to this section when you are stress-testing the system

better code:

• [FEAT]: refactor llm_engine.py #5
• [FEAT]: improve typing quality #4

better UI:

• Improve the episode rendering function #3

Scenario and evaluation schema:

We need to find a better structure to support a more accurate evaluation

Populate scenarios:

Once we figure out the schema, we should start populating scenarios.