AI/ML Development

Build, debug, and optimize Claude API / Anthropic SDK apps. Covers prompt caching, model migrations (4.5→4.6→4.7), tool use, thinking, batch, files, citations, and memory.

Generate and edit images using OpenAI's Image API for visual projects and prototyping.

HF Hub CLI for managing models, datasets, repos, and compute jobs on Hugging Face.

Train models with TRL: SFT, DPO, GRPO fine-tuning and GGUF conversion on Hugging Face.

Create and manage datasets with configs and SQL querying on the Hugging Face Hub.

Model evaluation with vLLM/lighteval benchmarking and eval tables on Hugging Face.

Track ML experiments with real-time dashboards and metrics logging on Hugging Face.

77 AI research skills across 20 categories covering training, inference, and MLOps workflows.

AI and ML skills for .NET 8+: technology selection (ML.NET, Microsoft.Extensions.AI, Microsoft Agent Framework, Copilot SDK, ONNX, OllamaSharp), LLM integration, agentic workflows, RAG pipelines, MCP server creation.

Expert guidance for Microsoft Foundry (Azure AI Foundry): building agents with Agent 365, MCP tools, Azure OpenAI, VNets, guardrails, and CI/CD. Requires the mcp_microsoftdocs MCP for live docs.

Expert guidance for the classic Microsoft Foundry experience (legacy Azure AI Studio workflows). Requires the mcp_microsoftdocs MCP for live docs.

Run Microsoft Foundry locally for offline / on-device AI development. Requires the mcp_microsoftdocs MCP for live docs.

Tooling and CLI patterns for Microsoft Foundry: deployment helpers, scripts, and integration utilities. Requires the mcp_microsoftdocs MCP for live docs.

Train, deploy, and monitor ML models on Azure ML: workspaces, compute, pipelines, prompt flow, and managed endpoints. Requires the mcp_microsoftdocs MCP for live docs.

Build search and RAG indexes on Azure Cognitive Search: vector + hybrid search, semantic ranking, indexers, and skillsets. Requires the mcp_microsoftdocs MCP for live docs.