diff --git a/fern/docs.yml b/fern/docs.yml index d824b1dcc..0618cdada 100644 --- a/fern/docs.yml +++ b/fern/docs.yml @@ -216,10 +216,15 @@ redirects: destination: "/nemo/gym/main/build-verifiers" - source: "/nemo/gym/main/build-verifiers/resources-server" destination: "/nemo/gym/main/build-verifiers" + # Example Resources Servers page was removed; fold into the Build Verifiers index + - source: "/nemo/gym/build-verifiers/example-resources-servers" + destination: "/nemo/gym/main/build-verifiers" + - source: "/nemo/gym/main/build-verifiers/example-resources-servers" + destination: "/nemo/gym/main/build-verifiers" - source: "/nemo/gym/resources-server/example-resources-servers" - destination: "/nemo/gym/main/build-verifiers/example-resources-servers" + destination: "/nemo/gym/main/build-verifiers" - source: "/nemo/gym/main/resources-server/example-resources-servers" - destination: "/nemo/gym/main/build-verifiers/example-resources-servers" + destination: "/nemo/gym/main/build-verifiers" # Version alias: latest → main (the `Latest` slug was retired; trunk lives at `main`) # Intentionally last so .html stripping and specific reorg rules win first. - source: "/nemo/gym/latest" diff --git a/fern/versions/latest/pages/build-verifiers/example-resources-servers.mdx b/fern/versions/latest/pages/build-verifiers/example-resources-servers.mdx deleted file mode 100644 index da2065c63..000000000 --- a/fern/versions/latest/pages/build-verifiers/example-resources-servers.mdx +++ /dev/null @@ -1,35 +0,0 @@ ---- -title: "Example Resources Servers" -description: "Concrete Resources Servers and the task, action, and verification patterns they demonstrate." -position: 3 ---- - -Resources Servers can model anything from a stateful workplace simulation to an isolated code execution sandbox. These examples show the shape of real server implementations and how task setup, tool actions, and verification fit together. - -## `workplace_assistant` - -[`workplace_assistant`](https://github.com/NVIDIA-NeMo/Gym/tree/main/resources_servers/workplace_assistant) implements multi-step tool calling in a workplace setting. - -- **Task**: Execute business activities such as sending emails, scheduling meetings, and managing projects. -- **Actions**: 26 tools across 5 databases: email, calendar, analytics, project management, and CRM. Each tool can read and mutate the database state. -- **Verification**: State matching: executes both the agent's actions and the ground truth actions against fresh databases, then compares the resulting states. - -## `math_with_code` - -[`math_with_code`](https://github.com/NVIDIA-NeMo/Gym/tree/main/resources_servers/math_with_code) implements mathematical reasoning with code execution. - -- **Task**: Solve math problems using Python as a reasoning tool. -- **Actions**: `execute_python()` runs code in an isolated per-session process with numpy, scipy, and pandas available. State persists across steps so the agent can build on previous computations. -- **Verification**: Answer correctness: extracts the boxed answer from the model's final response and compares it against the expected result. - - - - -Review the Resources Server role, state model, tool endpoints, and verifier interface. - - - -Build a minimal environment with a custom Resources Server. - - - diff --git a/fern/versions/latest/pages/build-verifiers/multi-reward-verification.mdx b/fern/versions/latest/pages/build-verifiers/multi-reward-verification.mdx index 456c67008..6674f53b9 100644 --- a/fern/versions/latest/pages/build-verifiers/multi-reward-verification.mdx +++ b/fern/versions/latest/pages/build-verifiers/multi-reward-verification.mdx @@ -1,7 +1,7 @@ --- title: "Multi-Reward Verification" description: "Score a rollout on multiple reward components for richer evaluation and multi-objective RL" -position: 2 +position: 3 --- diff --git a/fern/versions/latest/pages/build-verifiers/verification-patterns/index.mdx b/fern/versions/latest/pages/build-verifiers/verification-patterns/index.mdx index b60e54e20..bb83ff6d8 100644 --- a/fern/versions/latest/pages/build-verifiers/verification-patterns/index.mdx +++ b/fern/versions/latest/pages/build-verifiers/verification-patterns/index.mdx @@ -1,7 +1,7 @@ --- title: "Verification Patterns" description: "Patterns for evaluating agent behavior and computing scores." -position: 4 +position: 2 --- Verification is how an environment evaluates agent behavior and computes a score. Every environment implements some form of verification — the pattern you choose depends on your task.