Chris Wing

## Summary Set up PyPI publishing so users can install NeMo Gym via `pip install nemo-gym`. ## Requirements - [ ] Create PyPI project for `nemo-gym` - [ ] Configure...

### Problem The `resources_servers/simple_weather/README.md` is essentially empty (only licensing info). This is particularly problematic because simple_weather is the **primary introductory example** featured in: - Main repository README - Tutorial 01...

### Background BrowseComp (Browser-based Composition) is a benchmark for evaluating agents on web search and information gathering tasks. Users have requested integration of this benchmark in NeMo Gym ### What...

### Background SWE-Bench Verified is a popular benchmark for evaluating agents on real-world software engineering tasks. ### What is SWE-Bench Verified SWE-Bench Verified contains GitHub issues from open-source projects where...

### Background Tau Bench is a popular benchmark for evaluating agentic tool use capabilities. Multiple users have requested this environment. ### What is Tau Bench Tau Bench evaluates agents on...

## Add Deployment Topology Documentation ### Background Users have asked whether NeMo Gym runs in the same cluster as NeMo RL, suggesting confusion about the physical deployment model and what...

### Background Users have asked if they could "use Gym with external models for DPO data collection" as if this was a special case, when it's actually a core use...

## Clarify Task/Tool Mental Model and When to Create New Resource Servers ### Background Users have asked "do you need another resource server if you have different tool sets?" and...

## Create Database Integration Tutorial ### Background Multiple users have asked how to handle database operations in NeMo Gym for stateful agentic tasks. This is a common need for agentic...

Referenced the workplace assistant example's verify function here: https://github.com/NVIDIA-NeMo/Gym/blob/main/resources_servers/workplace_assistant/app.py#L99 . This verify basically resets the states and executes the predicted functions and ground truth functions. How would we pull the...