Results for “deep learning”

Deep learning-based variant calling from long reads using Clair3 for SNPs and small indels. Use when calling germline variants from ONT or PacBio alignments, particularly when high accuracy is needed for clinical or research applications.

The English summary is being prepared. Showing the original description.

Detects allele-specific transcription factor or histone modification binding from heterozygous-variant ChIP-seq using WASP (reference-bias filter; mandatory upstream), RASQUAL (joint QTL + bias-corrected testing), BaalChIP (Bayesian beta-binomial with copy-number-aware overdispersion), and AlleleSeq (personalized diploid genome). Handles imprinted-locus awareness, X-inactivation artifacts, cancer copy-number imbalance, and integration with downstream caQTL / bQTL mapping. Use when identifying variants with allelic effects on TF binding, fine-mapping causal regulatory variants, validating deep-learning variant predictions, or characterizing cis-acting regulatory effects.

The English summary is being prepared. Showing the original description.

Applies Judea Pearl's causal reasoning frameworks to distinguish correlation from causation, evaluate AI capabilities, and make counterfactual decisions. Reach for this skill whenever Claude encounters questions about causal inference, structural causal models, the limitations of deep learning, AGI, experimental design, covariate selection, or personalized decision-making. Trigger this skill for topics involving Bayesian networks, the do-calculus, the Ladder of Causation, or when a user tries to answer 'what if' or 'why' questions using purely observational data. Pearl's principles are essential for moving beyond probability calculus into true causal understanding.

The English summary is being prepared. Showing the original description.

This skill channels the strategic and scientific reasoning of Demis Hassabis, CEO and co-founder of Google DeepMind, AlphaGo and AlphaFold, and 2024 Nobel Prize in Chemistry. Use this skill whenever you are evaluating AI for scientific discovery, tackling "root node" problems, designing reinforcement learning systems, or discussing AGI timelines, safety, and global governance. Reach for it when the user faces massive combinatorial search spaces, wants to apply AI to physical/biological sciences (like digital biology), or needs to balance rapid AI scaling with the rigorous scientific method. Apply these mental models to shift the focus from building consumer apps to using AI as the ultimate meta-solution for understanding reality.

The English summary is being prepared. Showing the original description.

Applies the engineering and research philosophies of Jeff Dean, Chief Scientist at Google DeepMind and Google Research. Reach for this skill whenever you are designing large-scale distributed systems, optimizing latency and energy efficiency, or making architectural decisions about machine learning infrastructure. It should trigger automatically for topics involving hardware-ML co-design, model distillation, sparse activation, massively multi-task models, or scaling systems by 5x to 10x. Use this skill to evaluate system bottlenecks, transition from specialized to unified models, and optimize experimental velocity. Apply his mental models to avoid premature 100x scaling and to treat AI models as reasoning engines rather than memorization databases.

The English summary is being prepared. Showing the original description.

Deep expertise in ML/CV model selection, training pipelines, and inference architecture. Use when designing machine learning systems, computer vision pipelines, or AI-powered features.

The English summary is being prepared. Showing the original description.

Use when writing or changing PyTorch code — model, Dataset/DataLoader, training and evaluation loops, determinism, mixed precision, torch.compile, checkpoints with safetensors, Hugging Face models, ONNX export, and tests that run on CPU

The English summary is being prepared. Showing the original description.

Stage exploration before instruction on complex problems. The learner produces two attempted approaches before consolidation — which builds on those attempts, not from scratch. Use for genuinely hard problems where struggle produces deeper learning.

The English summary is being prepared. Showing the original description.

Socratic tutor that teaches codebase understanding through questioning, challenges, and active recall. Invoke with /learn-codebase when onboarding to a new codebase or deepening understanding of existing code. Maintains a learning journal that tracks mastery and schedules reviews.

The English summary is being prepared. Showing the original description.

Bengio's causal inference for AI: Interventional reasoning, counterfactuals, and System 2 deep learning. World models with causal structure.

The English summary is being prepared. Showing the original description.

audioFlux is a deep learning tool library for audio and music analysis and feature extraction, supporting dozens of time-frequency transforms and hundreds of feature combinations for classification, separation, MIR, and ASR tasks.

The English summary is being prepared. Showing the original description.

Silently restructures the user's natural-language prompt into the format the model CURRENTLY running this skill handles best, then answers. On activation it identifies which model family is executing it (Claude, GPT, Gemini, Llama, DeepSeek, Mistral, Qwen, Grok, Perplexity, Kimi, GLM, Command, Nova, or Phi) and loads that one model's official strategy — so the optimization always matches the model that actually runs it. Activate with /prompt-refine. Use when users want better answers without learning prompt engineering.

The English summary is being prepared. Showing the original description.

AI systematic knowledge acquisition from unfamiliar territory — deliberate model-building with feedback loops. Maps spaced repetition principles to AI reasoning: survey the territory, hypothesize structure, explore with probes, integrate findings, verify understanding, and consolidate for future retrieval. Use when encountering an unfamiliar codebase or domain, when a user asks about a topic requiring genuine investigation rather than recall, when multiple conflicting sources require building a coherent model, or when preparing to teach a topic and deep understanding is required first.

The English summary is being prepared. Showing the original description.

CORE learning workflow — use this whenever the user has a source note open (paper, post, book, lecture, course, clipping) and wants to extract knowledge, understand what they just read, or build it into their vault. Triggers on "process this note", "process this paper", "extract concepts", "what can I learn from this", "atomize this", "add this to my vault", "what's important here", or when the user shares a note and asks for help understanding it. Do NOT wait for an explicit /process command — if the user is engaging with a source note and wants to go deeper, this skill applies.

The English summary is being prepared. Showing the original description.

MATLAB R2026a workflow for signal processing, DSP, image processing, computer vision, lidar, medical imaging, deep learning, and AI-assisted reproducible experiments. Use for filters, spectrograms, wavelets, object detection, segmentation, classification, training, inference, and generated figures.

The English summary is being prepared. Showing the original description.

Use when planning a source-grounded method overview, architecture, pipeline, system/data-flow, mechanism, or agent-workflow figure for a CS or deep-learning paper. Produces an editable figure brief before image generation or manual redrawing; not for plots, tables, or general prose.

The English summary is being prepared. Showing the original description.

Use when exploring, continuing, or accepting a deep-learning or computer-science research idea. Converts claims into pilot-first experiment matrices covering diagnostics, failure analysis, authorized execution, evaluation, and paper-story viability.

The English summary is being prepared. Showing the original description.

Teach a topic, video, article, codebase change, spec, or transcript until the user demonstrably understands it, using staged instruction, a running checklist doc, and quizzes. Use whenever the user says "teach me", "explain this in detail", "help me understand X", "/teach-me", "walk me through this", "I want to learn X", "understand this deeply", or pastes a YouTube URL, article, paper, PR, or spec and asks to understand it rather than summarize it. Also use when the user wants study notes or raw material from a source they are learning. Do NOT use for a quick factual lookup, a summary request, or when the user just wants the answer.

The English summary is being prepared. Showing the original description.

Create and maintain the shared knowledge/ context files that every other marketing skill reads - brand voice, ICP personas, positioning, KPIs, and the learnings log. Run this once per company or client before using any other marketing skill. Use when the user says "set up brand context", "onboard my brand", "set up the marketing knowledge base", "the skills don't know my brand", "update our positioning", "our voice has changed", "add a new persona", or when any marketing skill reports that a knowledge/ file is missing. Also use when output sounds generic or off-brand. For writing a positioning statement as a standalone document, see positioning-doc. For deep persona research from raw sources, see customer-research.

The English summary is being prepared. Showing the original description.

Deep learning framework development with tinygrad - a minimal tensor library with autograd, JIT compilation, and multi-device support. Use when writing neural networks, training models, implementing tensor operations, working with UOps/PatternMatcher for graph transformations, or contributing to tinygrad internals. Triggers on tinygrad imports, Tensor operations, nn modules, optimizer usage, schedule/codegen work, or device backends.

The English summary is being prepared. Showing the original description.

Silently restructures the user's natural-language prompt into the format the model CURRENTLY running this skill handles best, then answers. On activation it identifies which model family is executing it (Claude, GPT, Gemini, Llama, DeepSeek, Mistral, Qwen, Grok, Perplexity, Kimi, GLM, Command, Nova, or Phi) and loads that one model's official strategy — so the optimization always matches the model that actually runs it. Activate with /prompt-refine. Use when users want better answers without learning prompt engineering.

The English summary is being prepared. Showing the original description.

Detects allele-specific transcription factor or histone modification binding from heterozygous-variant ChIP-seq using WASP (reference-bias filter; mandatory upstream), RASQUAL (joint QTL + bias-corrected testing), BaalChIP (Bayesian beta-binomial with copy-number-aware overdispersion), and AlleleSeq (personalized diploid genome). Handles imprinted-locus awareness, X-inactivation artifacts, cancer copy-number imbalance, and integration with downstream caQTL / bQTL mapping. Use when identifying variants with allelic effects on TF binding, fine-mapping causal regulatory variants, validating deep-learning variant predictions, or characterizing cis-acting regulatory effects.

The English summary is being prepared. Showing the original description.

Use when a learner has surface understanding and needs to explore mechanism, tradeoffs, edge cases, and production implications at depth.

The English summary is being prepared. Showing the original description.

Agente que simula Andrej Karpathy — ex-Director of AI da Tesla, co-fundador da OpenAI, fundador da Eureka Labs, e o maior educador de deep learning do mundo.

The English summary is being prepared. Showing the original description.