Data analysis, machine learning, LLM apps, and prompt design (9.1K skills)
Set up reusable AWS Glue connections to databases, Redshift, Snowflake, and BigQuery. Discover existing connections, configure credentials and networking, then test access through Glue and the intended processing engine to diagnose connection problems.
Design relationship-based applications with Amazon Neptune, from fraud detection and recommendations to knowledge graphs. Get guidance on engine selection, graph modeling, queries, Neo4j migration, and troubleshooting, with resource creation and supported changes requiring explicit confirmation.
Plan migrations and configure Amazon OpenSearch Service or Serverless for search and analytics. Explore semantic search, which finds meaning beyond keywords, analyze logs and distributed traces, and investigate incidents through natural language questions.
Inventory AWS Glue Data Catalog assets across S3 Tables, Redshift-federated catalogs, and remote Iceberg catalogs. Summarize databases, tables, storage locations, formats, and schemas, and flag unregistered or stale tables without running queries.
Find AWS datasets using a table name, business keyword, column name, or S3 path. Search Glue Data Catalog, S3, S3 Tables, and Redshift to return matching table references, data formats, and storage locations.
Run SQL queries with Amazon Athena across Glue, S3 Tables, and Redshift catalogs. Inspect table contents, select query execution settings, and review results alongside query costs, scanned data, and execution time.
Write SQL and troubleshoot Amazon Redshift using its own behavior rather than PostgreSQL assumptions. Covers slow queries, connection and permission errors, data loading, storage growth, and operations for both provisioned clusters and Serverless environments.
Build and troubleshoot Cloudflare Basin workflows that collect and transform events, store them in R2 tables, manage table metadata, and query data with SQL. Covers Basin Pipelines, Basin Catalog, and Basin SQL.
Query and analyze logs, device telemetry, and time series in Azure Data Explorer using KQL, its query language. Explore table structures, locate data across candidate sources, summarize events, and investigate trends or related errors.
Helps you use Azure AI services for keyword and vector search, which finds similar meanings, speech transcription, spoken audio generation, and text extraction from documents. Includes service operations and SDK references for implementation.
Build and deploy AI agents and models in Microsoft Foundry, then evaluate responses, improve instructions, and fine-tune models. The skill also guides troubleshooting, access permissions, and capacity management throughout development and operation.
Make local or Azure Blob documents searchable through Foundry IQ knowledge bases. The skill guides setup, retrieval with source citations, connections to existing agents, and diagnosis of configuration problems or unsupported source requests.
Deploy Azure OpenAI models using defaults or custom settings for version, deployment type, capacity, and content filtering. Check model support and available quota, meaning the subscription's usage allowance, to identify suitable regions and projects.
Fine-tune models on Microsoft Foundry with guidance for choosing a training method, preparing and validating datasets, submitting and monitoring jobs, and evaluating and deploying trained models. Compare results with the original model and track token costs.
Implement document processing in TypeScript with Azure Document Intelligence. Extract text, tables, and fields from invoices, receipts, and IDs, or build custom models and classifiers to handle specialized documents and distinguish document types.
Build AI applications in TypeScript with the Azure AI Projects SDK. The skill guides agent creation and execution, OpenAI client access, and management of connected resources, model deployments, datasets, and search indexes.
Build real-time voice assistants in JavaScript or TypeScript with Azure AI. Covers two-way audio streaming, text responses, speech boundary detection, and calling application functions during conversations in Node.js or browsers.
Build document search features in Python with Azure AI Search. Create indexes, upload documents, combine keyword matching with vector search that finds similar meaning, and configure semantic ranking, filters, and search suggestions.
Build AI applications on Microsoft Foundry with Python. Create and version agents, connect tools, run response evaluations, and manage project resources such as connections and datasets through the SDK or an OpenAI-compatible client.
Manage Azure Machine Learning resources with Python, from registering data and trained models to creating compute clusters and running training jobs. Includes patterns for pipelines that connect multiple processing steps and reusable training environments.
Helps implement Python workflows that extract information from documents, images, audio, and video using Azure AI Content Understanding. Covers document text, timed transcripts, video frames, and custom invoice fields for search and automated processing.
Build Python features that identify user intent and extract information such as names from conversation text using Azure AI Language. Includes guidance on authentication, request structure, error handling, and proper cleanup of client resources.
Implement text analysis in Python with Azure AI Language. Covers sentiment detection, extracting names and key phrases, identifying languages, masking personal information, and extracting medical terms, with examples for batch processing and asynchronous clients.
Add text and image moderation to Python applications using Azure AI Content Safety. Analyze harmful content by category and severity, manage custom text blocklists, and use the results to apply thresholds for user submissions or AI outputs.