Migrate existing NeMo Flow code and documentation to NeMo Relay by updating explicit names, imports, paths, and configuration references. A bundled helper previews changes, while language-specific checks cover dependencies, generated files, and remaining legacy identifiers.
Tune adaptive behavior in an instrumented NeMo Relay application using runtime data. Configure tool parallelism, prompt caching, and state storage, then compare changes against a baseline to check latency, correctness, and failure rates.
Export existing medical AI inference results to MLflow, a system for recording machine learning runs. Preview sanitized metadata, parameters, reproducibility details, and engineering quality metrics before export, with optional image artifacts under explicit upload policies.
Plan and troubleshoot Nemotron retrieval recipes for embeddings, which turn text into searchable vectors, and reranking, which reorders search results. Review configurations, prepare training and evaluation commands, and check export or deployment requirements.
Run training with mock or real data in Megatron-LM and Megatron Bridge, match their settings, and compare loss curves, which show training error over time. Includes recipes and multi-GPU examples for checking training behavior.
Preserve coding-agent progress in readable session files, including goals, decisions, changed files, and next steps. Helps resume long-running work after disconnects or editor restarts and gives another agent concise handoff instructions.
Try NeMo Relay through a CLI wrapper, a supported integration, or direct Python, Node.js, or Rust instrumentation. Choose a minimal first run and verify recorded tool and model activity before considering one additional plugin.
Guide repeated NeMo-RL or NeMo-gym experiments to test hypotheses and improve training settings. Establish a baseline, compare relevant metrics, and preserve reproducible records in git and TSV logs while keeping users involved in launch decisions.
Package reusable NeMo Relay behavior as an embedded component or an independently distributed plugin. Design shared configuration, validate it before activation, and register behavior with rollback handling so failed setup does not leave partial changes active.
Guides typed data integration in NeMo Relay by choosing codecs, converters between application values and JSON. Helps preserve middleware edits and caller-visible responses while distinguishing application data conversion from provider-specific LLM request and response handling.
Choose Megatron Bridge training configurations for a model, GPU allocation, and training goal. Get starting commands and parallel processing adjustments, with clear distinctions between changes that affect learning behavior, execution tuning, and benchmark-only settings.
Convert single-node training or inference commands into multi-node Slurm batch jobs. Choose an appropriate process launch method, configure shared storage and containers, and investigate communication timeouts, launch errors, and GPU memory shortages.
Search documents through NeMo Retriever MCP and use retrieved passages to support answers. Add missing documents when server access is confirmed, then search them while preserving available document identifiers and page numbers for citations.
Choose how to split large-model training across GPUs in Megatron Bridge. This guide helps combine parallelism methods, estimate GPU requirements, match settings to hardware connections, and check memory use and performance after configuration changes.
Choose and install the right NeMo Relay option for a command-line tool, a Python, Node.js, or Rust application, or a supported framework integration. Check basic availability and report the installation result before moving to runtime setup.
Turn rough moderation requirements or existing policies into custom safety rules for NVIDIA Nemotron models. Produce a Markdown policy, a JSON category structure, and ready-to-use classification prompts with definitions, exceptions, examples, and rules for keeping conversations on topic.
Enable Megatron FSDP in Megatron-Bridge to distribute model training state across GPUs. Follow configuration and checkpoint requirements, investigate memory failures or performance regressions, and verify the setup with an existing two-GPU functional test.
Plan and configure Nemotron customization workflows using existing steps for data preparation, translation, training, conversion, optimization, and evaluation. Check how outputs connect between stages and produce configuration files suited to the supplied models, data, and hardware.
Keep NeMo Relay event histories separate for concurrent requests, tasks, and agents. The skill guides independent scope stacks, which track nested operations, and preserves parent-child relationships when execution crosses asynchronous tasks, threads, or workers.
Find starting configurations for training MoE models, which route work among specialized expert networks, on H100, B200, GB200, and GB300. Provides model-specific parallel layouts, communication options, and approximate performance bands to guide measured comparisons.
Helps configure training across multiple GPUs or machines in NeMo AutoModel. Guides strategy selection, YAML settings, and parallelism choices that split model computation or data, with constraints for large models, long sequences, and expert models.
Plan and tune training for vision-language models that use MoE, a mixture of expert networks, in Megatron Bridge. Compares FSDP and 3D parallel training, with guidance on memory use, batch sizes, and performance checks using realistic image inputs.
Guides hierarchical context parallelism in Megatron-Bridge, which splits input sequences across nested GPU groups. Covers configuration, implementation locations, constraints, and log checks to help enable the feature and investigate memory errors or performance regressions.
Build, migrate, and review adapters that connect external agent runtimes to NVIDIA NeMo Fabric. It guides configuration translation, startup and shutdown, discovery metadata, and validation against the public contract, the rules defining how adapters interact with Fabric.
Is this your account? Link GitHub to bring it onto your creator page (Japanese)