Loading...
Loading...
Found 31 Skills
Stage 4 of the Clinical ASR Flywheel. Use when priority KER is above 0.3 to run stock NeMo SFT on Parakeet TDT v2 and offline cycle N+1 re-eval. NOT for generic word boosting (use /finetune-asr).
Fine-tune LLMs with Unsloth using GRPO or SFT. Supports FP8, vision models, mobile deployment, Docker, packing, GGUF export. Use when: train with GRPO, fine-tune, reward functions, SFT training, FP8 training, vision fine-tuning, phone deployment, docker training, packing, export to GGUF.
Implements the Syncfusion WPF SfTextBoxExt autocomplete features for intelligent suggestions, filtering, and tokenized multi-selection. Use when adding predictive text input, filtered suggestion lists, or multi-select search boxes.
Cosmos-Reason2-8B video QA supervised fine-tuning with FSDP parallelism. Use when training or evaluating video question-answering models, fine-tuning Cosmos-Reason2 with SFT, or working with Cosmos-RL. Trigger phrases include "fine-tune Cosmos-Reason", "Cosmos-RL SFT", "video QA fine-tune", "Cosmos-Reason2-8B training".
Create and customize WPF time picker controls using Syncfusion's SfTimePicker component. Use this skill whenever the user needs to implement time selection UI, handle time input values, format time display, customize time selector behavior, or style time picker controls. Triggers on: time picker, SfTimePicker, time selection, time input field, time formatting, time value binding, time selector customization, WPF time control, time editing, hourly/minute selection.
Implement Syncfusion WPF SfTreeNavigator for in-place hierarchical tree navigation and drill-down panels. Use this when building hierarchical navigation panels, drill-down navigation, or breadcrumb-style tree navigation in WPF. Covers declarative item creation, ItemsSource data binding with HierarchicalDataTemplate, Default and Extended navigation modes, HeaderTemplate customization, and MVVM-based SelectedItem selection.
Guide for implementing Syncfusion Windows Forms Tooltip (SfToolTip) component for displaying contextual information and hover text in Windows Forms applications. Use this skill when implementing tooltips, popup information, balloon tips, or informational popups on controls. Covers SfToolTip setup, multi-item tooltips, custom tooltip content, appearance customization, and tooltip images.
Curated research collection on adaptation strategies for agentic AI systems, covering agent and tool adaptation methods with RL, SFT, and DPO approaches
This skill should be used when users want to train or fine-tune language models using TRL (Transformer Reinforcement Learning) on Hugging Face Jobs infrastructure. Covers SFT, DPO, GRPO and reward modeling training methods, plus GGUF conversion for local deployment. Includes guidance on the TRL Jobs package, UV scripts with PEP 723 format, dataset preparation and validation, hardware selection, cost estimation, Trackio monitoring, Hub authentication, and model persistence. Should be invoked for tasks involving cloud GPU training, GGUF conversion, or when users mention training on Hugging Face Jobs without local GPU setup.
Validate and use packed sequences and long-context training in Megatron-Bridge, distinguishing offline packed SFT for LLMs from in-batch packing for VLMs, and applying the right CP constraints.
Train and fine-tune transformer language models using TRL (Transformers Reinforcement Learning). Supports SFT, DPO, GRPO, KTO, RLOO and Reward Model training via CLI commands.
Plan, configure, and chain repo-native Nemotron customization steps into single-step or multi-step pipelines: curation, translation, SFT/PEFT (AutoModel or Megatron-Bridge), pretraining/CPT, RL alignment (DPO/RLVR/GRPO/RLHF), BYOB/MCQ benchmarks, checkpoint conversion, ModelOpt optimization, env profiles, and evaluation of trained checkpoints or existing/hosted endpoints. Use when a request names a Nemotron step or workflow, or asks to clean, translate, train, fine-tune, align, convert, optimize, evaluate, or compose these into a pipeline. Do NOT use for frontend/dashboard/visualization work, generic ML advice, billing/access, or non-Nemotron coding tasks.