Find the perfect skill

Data & AI

Scaffold matplotlib figure code following established TDA visualization conventions. Ensures publication-ready figures with consistent styling and save patterns.

Data & AI

Generate publication-ready statistical tables using gt, kableExtra, or flextable. Covers descriptive statistics, regression results, ANOVA tables, correlation matrices, and APA formatting.

Data & AI

Performs Bayesian inference and probabilistic modeling for geospatial data. Use it to build hierarchical models, compute posteriors via MCMC or variational methods (PyMC/TFP), compare models (LOO/WAIC/DIC), or implement spatial Gaussian processes with Cholesky decomposition.

Data & AI

Run the full skill evolution pipeline: harvest sessions, discover signals, build golden dataset, evaluate baseline, evolve via DSPy, and compare before/after scores.

Data & AI

Searches Reddit, X/Twitter, and the web for genuine recent opinions and sentiment on any topic, producing a structured report with consensus, pain points, and signal confidence.

Data & AI

Build scalable data pipelines, modern data warehouses, and real-time streaming architectures. Master the modern data stack including Spark, dbt, Airflow, and cloud-native platforms.

Data & AI

Analyze text using traditional NLP and LLM methods. Extract sentiment, topics, and insights from various content types.

Data & AI

Classifies stablecoins to override default resilience sub-factor scores (chainRisk, collateralQuality, custodyModel) when the default inference is incorrect. Useful after adding new stablecoins or auditing existing scores for accuracy.

Data & AI

Automates the full workflow for Boltz-2 protein structure prediction: upload, spec generation, validation, job submission, and status tracking. Supports nanobody-target cross-model workflows.

Data & AI

Debug PyTorch 2 compiler stack failures including Dynamo graph breaks, Inductor codegen errors, AOTAutograd crashes, and accuracy mismatches.

Data & AI

Investigates BGP announcements, prefix ownership, and AS paths to identify the organization and country for an AS missing from overlay.json. It helps when an AS has active BGP prefixes but no published metadata, enabling accurate network attribution.

Data & AI

Build and apply lead scoring models combining firmographic fit, behavioral signals, and intent data for sales prioritization.