Skill Creator
ScientiaCapital
12 published skills · 8 GitHub stars across the catalog. View on GitHub ↗
unsloth-tokenizer
Analyze, compare, and work with tokenizers using Unsloth tools. Compare different tokenizers, analyze token efficiency, and integrate with Unsloth models. For SuperBPE training, see the 'superbpe' skill.
training-optimization
Advanced techniques for optimizing LLM fine-tuning. Covers learning rates, LoRA configuration, batch sizes, gradient strategies, hyperparameter tuning, and monitoring. Use when fine-tuning models for best performance.
superbpe
Train and use SuperBPE tokenizers for 20-33% token reduction across any project. Covers training, optimization, validation, and integration with any LLM framework. Use when you need efficient tokenization, want to reduce API costs, or maximize context windows.
adaptive-workflows
Self-learning workflow system that tracks what works best for your use cases. Records experiment results, suggests optimizations, creates custom templates, and builds a personal knowledge base. Use to learn from experience and optimize your LLM workflows over time.
unsloth-finetuning
Fine-tune LLMs 2x faster with 80% less memory using Unsloth. Use when the user wants to fine-tune models like Llama, Mistral, Phi, or Gemma. Handles model loading, LoRA configuration, training, and model export.
model-deployment
Export and deploy fine-tuned models to production. Covers GGUF/Ollama, vLLM, HuggingFace Hub, Docker, quantization, and platform selection. Use after fine-tuning when you need to deploy models efficiently.
dataset-engineering
Create, clean, and optimize datasets for LLM fine-tuning. Covers formats (Alpaca, ShareGPT, ChatML), synthetic data generation, quality assessment, and augmentation. Use when preparing data for training.
unsloth-mcp-server
Work with the Unsloth MCP Server codebase. Use when maintaining, extending, or debugging this specific MCP server implementation. Provides architecture knowledge, code patterns, and development workflows.
cost-optimized-llm
Implement cost-optimized LLM routing with NO OpenAI. Use tiered model selection (DeepSeek, Haiku, Sonnet) to achieve 70-90% cost savings. Triggers on "LLM costs", "model selection", "cost optimization", "which model", "DeepSeek", "Claude pricing", "reduce AI costs".
deployment-patterns
Deploy projects to Vercel, Railway, or Docker with platform-specific best practices. Use when deploying applications, configuring deployment settings, debugging deployment failures, or setting up CI/CD pipelines. Triggers on "deploy to vercel", "railway deployment", "docker build", "deployment failed", "configure vercel.json".
revenue-acceleration
GTM workflows for revenue acceleration across Scientia projects. Use for demo preparation, sales outreach, battle cards, pricing strategy, and revenue tracking. Triggers on "revenue focus", "prepare demo", "sales outreach", "battle card", "GTM strategy", "pricing", "tier-1 projects".
supabase-mastery
Master Supabase patterns for migrations, RLS policies, pgvector, and authentication. Use when creating database schemas, writing migrations, implementing row-level security, setting up auth, or debugging Supabase issues. Triggers on "supabase migration", "RLS policy", "row level security", "pgvector", "supabase auth", "magic link".