3. Data, Science & AI

Processes massive documents (PDF, Word, etc.) beyond context limits, enabling search, summarization, data extraction, and chunking for AI applications.

Provides techniques and tools to interpret machine learning model decisions and communicate them effectively to stakeholders.

Manages risks in machine learning model deployment, including identification, assessment, and mitigation strategies for business applications.

Defines data contracts specifying schema, quality standards, and SLAs for seamless data exchange between producers and consumers.

Structures English into unambiguous control flow for AI sessions, enabling multi-agent orchestration without dependencies.

Enhances Claude's capabilities by providing additional assistance and functionality, making the AI more effective for users.

Guides schema registry implementation, compatibility checks, and governance for event-driven data architectures.

Orchestrates and governs multiple AI agents, ensuring safe, compliant operations through inter-agent communication, conflict resolution, and policy enforcement.

Comprehensive guide to Apache Kafka for real-time data streaming, covering topics, producers, consumers, and stream processing best practices.

Automates machine learning pipeline creation including feature engineering, model selection, and hyperparameter tuning using Auto-sklearn, H2O, TPOT, and AutoKeras.

Enables statistical comparison of machine learning models in production via A/B testing, multi-armed bandits, and traffic splitting for performance optimization.

Provides real-time dashboards, alerts, and time-series visualization for IoT device status and sensor data using WebSocket.

Automates document ingestion, content extraction, and structuring for RAG system integration from diverse sources.

Automates data ingestion, transformation, and loading using ETL/ELT patterns and Apache Airflow for scalable data pipelines.

Provides best practices for dbt modeling, testing, and production workflows in modern data transformation pipelines.

End-to-end workflow for developing, testing, and deploying edge AI applications from model training to production.

Provides structured playbooks for resolving data quality issues, data loss, corruption, and pipeline failures in data systems.

Comprehensive guide to enhancing RAG retrieval quality through chunking, embedding, reranking, and hybrid search techniques.

Step-by-step guide for implementing PaddleOCR to extract text, tables, and analyze document layouts across multiple languages.

Provides strategies and tools for evolving database schemas and transforming data during application development.

Integrates Google Gemini API for AI applications, supporting multimodal inputs, reasoning, and generation via Python/TypeScript SDKs.

Reference guide for configuring Claude Code subagents, enabling custom agent creation, agent type understanding, and Task tool extension with specialized agents.

Assesses machine learning model prediction reliability through confidence scoring and calibration techniques.

Compresses AI agents/skills with a 5-stage iterative process to reduce token count while preserving understanding quality, ensuring optimal performance.