Results for “experiment”

Pattern for creating new Colab notebooks. Trigger when: (1) Creating a new notebook for experiments, (2) Adding notebook-based functionality, (3) Agent or validation notebooks, (4) Any notebook that uses GPU training infrastructure.

Design and evaluate A/B tests with power, sample size, and robust metric interpretation. Use when: (1) planning controlled experiments, (2) reading p-values/effects, (3) sequential testing safeguards. NOT for: dark-pattern optimization.

Design MVPs, validated learning experiments, and pivot-or-persevere decisions using Build-Measure-Learn. Use when the user mentions "MVP scope", "validated learning", "pivot or persevere", "vanity metrics", "test assumptions", "innovation accounting", "build-measure-learn", or "minimum viable experiment". Also trigger when deciding what to include in a first version, measuring startup progress, or evaluating whether to change direction on a product bet. Covers innovation accounting and actionable metrics. For 5-day prototype testing, see design-sprint. For customer motivation analysis, see jobs-to-be-done.

Access the AlphaFold DB of 200M+ AI-PREDICTED protein structures — retrieve models by UniProt accession, download PDB/mmCIF files, and analyze prediction confidence metrics (pLDDT, PAE). Use when a UniProt ID needs a computationally predicted 3D structure or when no experimental structure exists, for homology modeling, protein engineering, or structure-based drug discovery; for EXPERIMENTALLY determined structures (X-ray, cryo-EM, NMR) prefer alterlab-pdb, and for protein sequences, annotations, or accession ID mapping prefer alterlab-uniprot instead. Part of the AlterLab Academic Skills suite.

Design MVPs, validated learning experiments, and pivot-or-persevere decisions using the Build-Measure-Learn loop and innovation accounting.

Train reinforcement learning models using torchforge, Meta's PyTorch-native RL library for scalable, algorithm-focused experimentation with GRPO, DAPO, and custom loss functions.

Creates a professional paid-advertising strategy covering objectives, economics, platform selection, campaign architecture, audiences, budget, creative, measurement, experiments, governance, rollout, and reporting.

Runs an autonomous optimization loop that iteratively improves an artifact against an objective and evaluator using Hypothesis Tree Refinement, with subagent executors in isolated git worktrees.

Runs a UI redesign experiment by creating three git worktree branches for modern minimal, bold creative, and classic refined design directions, then implements and compares each approach.

Google quantum computing framework. Use when targeting Google Quantum AI hardware, designing noise-aware circuits, or running quantum characterization experiments. Best for Google hardware, noise modeling, and low-level circuit design. For IBM hardware use qiskit; for quantum...

Google quantum computing framework. Use when targeting Google Quantum AI hardware, designing noise-aware circuits, or running quantum characterization experiments. Best for Google hardware, noise modeling, and low-level circuit design. For IBM hardware use qiskit; for quantum...

Turn a theory-first research idea into a proof-oriented research plan. Use after idea-creator-analysis, or when the user asks for theorem targets, assumptions, proof obligations, impossibility routes, analysis validation, or a non-experimental plan.

Get a deep critical review of research from GPT via Codex MCP. Use when user says "review my research", "help me review", "get external review", or wants critical feedback on research ideas, papers, or experimental results.

Submit protein sequences to the Adaptyv Bio Foundry for experimental characterization (binding, thermostability, expression, fluorescence) and retrieve results via API or Python SDK.

Use after PNAS reviews arrive to triage the decision, prioritize experiments, and draft a point-by-point response that is respectful, evidence-led, and honest about limits. Do not run before the main text is actually revised.

Use when creating a note, document, script or project that shouldn't outlive a known date — interview prep, a migration, a course, a time-boxed experiment — so it gets deleted on time instead of rotting in the repository.

Generate and rank theory-first or proof-oriented research ideas. Use when the user wants non-experimental research ideas, theoretical methods, proof programs, theorem candidates, impossibility results, convergence/sample-complexity analyses, or "analysis" variants of idea creation.

Provides practical guidance for training Mixture-of-Experts Vision-Language Models in Megatron Bridge, comparing FSDP and 3D-parallel approaches with lessons from recent multimodal experiments.

Process and analyze high-throughput sequencing data with deepTools for quality control, normalization, comparison, and publication-quality visualizations of ChIP-seq, RNA-seq, and ATAC-seq experiments.

Frame an epic as a testable hypothesis with target user, expected outcome, and validation method. Use when defining a major initiative before roadmap, discovery, or delivery planning.

Generate falsifiable trade strategy hypotheses from market data, trade logs, and journal snippets, with ranked hypothesis cards, experiment designs, kill criteria, and optional strategy.yaml export.

Embody the voice and methodology of Aaron T. Beck, founder of CBT, to guide users through cognitive restructuring techniques like thought records and behavioral experiments.

结构化业务数据分析：附件读取与口径核验、定向筛选、规则/阈值判定、指标异动归因、漏斗/留存/实验分析、经营复盘及可审计报告。当用户提供 Excel、CSV、PDF、图片或多份业务材料，要求查数、判异常、解释变化、比较方案或形成行动建议时使用。Use for evidence-grounded analysis of structured business data, including filtering, rule checks, reconciliation, diagnostics, experiments, and decision reports.

Provides guidance for interpreting and manipulating neural network internals using nnsight with optional NDIF remote execution. Use when needing to run interpretability experiments on massive models (70B+) without local GPU resources, or when working with any PyTorch architecture.

Provides guidance for interpreting and manipulating neural network internals using nnsight with optional NDIF remote execution. Use when needing to run interpretability experiments on massive models (70B+) without local GPU resources, or when working with any PyTorch architecture.

Calculate and model paid-media financial metrics including CPA, CPL, CPC, CPM, ROAS, MER, break-even targets, contribution margin, LTV:CAC, impression-share opportunity, budgets, forecasts, and experiment economics.

Diagnose paid-ad campaigns and draft or apply optimizations using evidence, financial constraints, and experiments. Supports budget reallocation, bid changes, creative rotation, and CPA/ROAS improvement.

Analyze product data and manage product tooling in PostHog. Use when the user wants product analytics or insights, HogQL/SQL queries, feature flags, experiments and A/B tests, error tracking, session replay, surveys, LLM analytics, dashboards, data warehouse, or PostHog documentation.

Orchestrate a full theory-first idea discovery pipeline. Use when the user wants the analysis counterpart of idea-discovery: literature context, proof-oriented idea generation, novelty checking, DeepSeek critique, and an analysis plan instead of an experiment-first workflow.

Submit one or more wandb-logged training/finetuning runs to the HPC scheduler. `WANDB_PROJECT` is fixed per repo (snake_case basename); `WANDB_RUN_GROUP` is picked per invocation. The training script must take the experiment/group name as a config key (e.g. Hydra `meta.experiment_name=<group>`); the skill passes it on the command line. The working tree is committed first so each run pins to a real SHA. Delegates SLURM/PBS templating to `cluster-instructions`. Use when the user asks to submit, queue, launch, or kick off a wandb training/finetuning job.

Writes Alan Hirsch's 8-week transformational course content, including video scripts, readings, case studies, reflection questions, and field experiments, in his voice with formation-over-information pedagogy.

Design and improve product user onboarding (first-time user experience) to drive activation and early retention. Produces an Onboarding & Activation Pack (aha moment spec, first 30 seconds + first mile plan, onboarding journey map, experiment backlog, measurement plan). Use for Growth teams.

Analyze A/B test results with statistical significance, sample size validation, confidence intervals, and ship/extend/stop recommendations. Use when evaluating experiment results, checking if a test reached significance, interpreting split test data, or deciding whether to ship a variant.

Build an Opportunity Solution Tree from outcomes to opportunities, solutions, and tests. Use when a stakeholder request needs problem framing before you decide what to build.

Provides guidance for training Mixture-of-Experts models with long context windows, covering context parallelism sizing, selective recomputation, dispatcher choices, and practical patterns from recent experiments.

Debug LLM applications using the Phoenix CLI: fetch traces, analyze errors, structure trace review with open and axial coding, inspect datasets, review experiments, and query the GraphQL API.