Loading...
Loading...
Found 13 Skills
High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2× faster than DeepSpeedChat with distributed architecture and GPU resource sharing.
Understanding Reinforcement Learning from Human Feedback (RLHF) for aligning language models. Use when learning about preference data, reward modeling, policy optimization, or direct alignment algorithms like DPO.
Guides product management for human data platforms—annotation and labeling products, workforce workflows, task design, quality systems (gold sets, adjudication, inter-annotator agreement), customer ML-team project delivery, contributor experience, and privacy-safe handling of human-generated training data. Use when prioritizing roadmap for labeling/RLHF/eval data platforms, writing PRDs for annotation or QA features, defining success metrics for throughput and quality, scoping enterprise customer workflows, or balancing cost-quality-speed tradeoffs—not for hands-on model training (data-scientist), warehouse/analytics pipelines (data-warehouse-engineer), generic BRD workshops without product lens (business-analyst), AI solution architecture for copilots (applied-ai-architect-commercial-enterprise), or control implementation for audits (compliance-engineer). UX flows: product-designer. Eval harnesses: prompt-engineer-agent-prompts-evals. Pricing/packaging for platform: product-management-monetization.
Use when implementing RL algorithms, training agents with rewards, or aligning LLMs with human feedback - covers policy gradients, PPO, Q-learning, RLHF, and GRPOUse when ", " mentioned.
Guidelines for creating high-quality datasets for LLM post-training (SFT/DPO/RLHF). Use when preparing data for fine-tuning, evaluating data quality, or designing data collection strategies.
Use when "training LLM", "finetuning", "RLHF", "distributed training", "DeepSpeed", "Accelerate", "PyTorch Lightning", "Ray Train", "TRL", "Unsloth", "LoRA training", "flash attention", "gradient checkpointing"
Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.
Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.
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.
Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with flexible infrastructure backends.
Plan Nemotron customization pipelines from repo steps: SFT, PEFT/LoRA, AutoModel vs Megatron-Bridge, DPO/RLVR/GRPO/RLHF, curate-then-translate, BYOB/MCQ benchmark prep or translation, checkpoint conversion, ModelOpt optimization, and endpoint or checkpoint evaluation.
Self-hosted ML coding practice platform with 68 problems covering Transformers, diffusion, RLHF, and more — instant browser feedback, no GPU required.