3. Data, Science & AI

Enables enterprise RAG solutions for policy assistants, knowledge bases, and multi-hop question answering using Foundry IQ and Azure AI Search.

Configures and manages Model Context Protocol (MCP) servers within VS Code for AI model integration, including setup, configuration, and troubleshooting.

Defines agent roles and orchestration patterns for multi-agent AI systems, requiring strict adherence to role-specific instructions before task execution.

Expert assistance with PostgreSQL database management, including advanced SQL, schema design, indexing, and performance tuning.

Creates structured Claude agents with YAML frontmatter, system prompts, and examples for defining AI behavior and customization.

Enables on-device machine learning for object detection and image classification using TensorFlow Lite in React Native applications.

Provides Svelte components for interactive data visualizations including charts, graphs, and responsive designs using D3 and Observable Plot.

AI tool analyzing language and context to verify claims made by public figures and identify potential misinformation.

Plans explainable AI requirements using SHAP, LIME, attention visualization, and regulatory compliance strategies for model transparency.

Guides structured planning of machine learning projects using CRISP-DM, TDSP, and MLOps methodologies with defined phase gates and deliverables.

Optimizes AI/ML costs by estimating token usage, managing context windows, and implementing batch processing and caching strategies.

Designs multi-agent AI systems with orchestration patterns, tooling, memory management, and error handling for autonomous agent workflows.

Systematically designs, tests, and optimizes AI prompts for improved model performance and consistency.

Selects optimal AI/ML models by evaluating capabilities, benchmarks, cost-performance tradeoffs, and deployment requirements.

Loads prior session context to maintain conversation continuity, ensuring seamless user interaction without repetition.

Optimizes LLM prompts using best practices in prompt engineering to enhance model responses and command effectiveness.

Designs retrieval-augmented generation (RAG) pipelines with optimized chunking, embedding, and context assembly strategies.

Automates pattern discovery from reflection episodes and consolidates successful approaches into reusable patterns and skills for AI systems.

Records feedback on pattern effectiveness and stores episodes to train recommendation systems, enabling pattern discovery through machine learning.

Evaluates AI models for fairness using demographic parity, equalized odds, and bias mitigation strategies.

Validates CSV file syntax for IP geofeed data against RFC 8805 standards.

Extracts text and metadata from PDF documents using the pypdf library for data processing and analysis.

Adapts AI behavior across task-specific modes including brainstorm, implement, and debug for optimized task execution.

Guides implementation of AI2AI learning services including personality evolution, compatibility calculations, and anonymous data exchange for peer learning features.