⚙️

AI EngineeringRust

Adaptive Tuning

Adaptive Tuning is an AI Engineering skill for Claude Code, published by NVIDIA in NeMo-Relay.

165 stars67 forkson NVIDIA/NeMo-RelayAdded 2026/09/08+23% in starsRepository updated 2026/09/07

View source on GitHub Download repo ZIP

agentsaiatifatofclaude-codecodexdeepagentshermes-agentlangchainlanggraphmiddlewarenvidiaobservabilityopenclawopeninferenceoptimizationotelruntimesecurity

Install in seconds

Install Adaptive Tuning

Copy Adaptive Tuning into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.

terminal

npx degit https://github.com/NVIDIA/NeMo-Relay/tree/main/skills/nemo-relay-plugin-adaptive-tuning ~/.claude/skills/nemo-relay-plugin-adaptive-tuning

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/NVIDIA/NeMo-Relay.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file: skills/nemo-relay-plugin-adaptive-tuning/SKILL.md in NVIDIA/NeMo-Relay
Installs to: ~/.claude/skills/nemo-relay-plugin-adaptive-tuning
Collection: One of 25 skills cataloged from this repository
Category: AI Engineering — 3670 skills

What Adaptive Tuning does

Adaptive Tuning configures NeMo Relay’s adaptive plugin to adjust telemetry, state, hints, tool parallelism, and rollout behavior from runtime signals. Use it after baseline instrumentation when you want to evaluate measured changes against known traffic.

Adaptive Tuning is cataloged under AI Engineering on DirSkills. Adaptive Tuning comes from a repository tagged agents, ai, atif, atof and claude-code.

Documentation

README

Excerpt

Tune Adaptive Plugin Behavior

Use This When

Use this skill when a user has baseline NeMo Relay instrumentation and wants to improve latency, parallelism, prompt-cache behavior, or model-request behavior from runtime signals. Keep adaptive behavior measured against a known baseline.

Do Not Use This When

Do not use this skill when the application is not instrumented yet. Start with nemo-relay-instrument-calls or nemo-relay-get-started first.

Default Guidance

This is the opening of the README. Read the full README on GitHub.

Frequently asked about Adaptive Tuning

What else does NVIDIA publish alongside Adaptive Tuning?
Adaptive Tuning is one of 25 skills that DirSkills catalogs from NVIDIA/NeMo-Relay, the repository it ships in. Its siblings there include Add Binding Feature, Add Middleware and Context Isolation. Each one is a separate skill with its own page in this directory, installs the same way Adaptive Tuning does, and is maintained by NVIDIA in that same repository. The rest of the collection is listed on the NVIDIA/NeMo-Relay page.
How does Adaptive Tuning compare to other AI Engineering skills?
Adaptive Tuning ranks #3356 by stars among the 3670 AI Engineering skills in this catalog. The most-starred ones next to it are Architecture Decision Records, AI-First Engineering and Agentic OS. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Adaptive Tuning against them. Open each page to compare what they document and how they install.

More from NVIDIA/NeMo-Relay

Adaptive Tuning is one of 25 skills cataloged on DirSkills from NVIDIA/NeMo-Relay.

See all 25 skills →

Add Binding Feature

Add Binding Feature adds or changes a public NeMo Relay API across the Rust core, FFI, and language bindings. Use it when a change must stay in parity across surfaces and needs coordinated validation.

Add Middleware

Add Middleware adds a new guardrail or intercept type to the NeMo Relay middleware pipeline. Use it when introducing a new registration surface or a middleware behavior at a new pipeline stage.

Context Isolation

Context Isolation helps keep NeMo Relay scope stacks separate across concurrent requests, async tasks, threads, workers, and agents. Use it when you need correct ancestry propagation and request-local middleware boundaries.

Contribute API

Contribute API helps add or change a public NeMo Relay API with binding parity, validation, and docs in mind. Use it when a runtime or binding surface changes and you need to keep affected bindings aligned.

Contribute Docs Or Examples

Contribute Docs Or Examples guides docs-only or example-heavy changes that match NeMo Relay’s public API and docs structure. Use it when updating user-facing docs, examples, or entry-point paths without changing implementation behavior.

Debug Runtime Integration

Debug Runtime Integration helps identify why NeMo Relay loads but runtime behavior is missing or incorrect, including scope, event, plugin, and export wiring issues. Use it when integration works partially but the expected callbacks or telemetry do not appear.

More AI Engineering skills

Adaptive Tuning is one of 3670 AI Engineering skills cataloged on DirSkills, ranked #3356 by stars among them.

Browse all AI Engineering skills →

Architecture Decision Records

Architecture Decision Records captures architectural decisions made during Claude Code sessions as structured ADRs, recording context, alternatives, and rationale. Use it when significant technical choices are made so future developers understand why the codebase is shaped the way it is.

AI-First Engineering

AI-First Engineering defines process, review, and architecture standards for teams where AI agents produce most implementation output. Use it to set ownership rules, quality gates, and evaluation criteria for agent-generated code.

Agentic OS

Agentic OS codifies a persistent multi-agent operating system architecture for Claude Code. Use it when building a multi-agent workflow with file-based memory, slash commands, and scheduled automation that survives session restarts.