Deep DiveAI Engineering
MAF v1 — Evaluation framework (Python + .NET)
Build an eval pipeline with golden datasets, scoring, smoke/full modes, and CI gates to catch prompt regressions before release.
Read article
Journal
Solutions Architect specializing in Azure, AI, and enterprise cloud-native development. Writing about architecture, AI agents, and building things that scale.
Deep DiveAI Engineering
MAF v1 — Python ↔ .NET asymmetries: a porting reference
Compare the six Python and .NET MAF API differences you will hit when porting samples, from constructors to workflow wiring.
Deep DiveAI Engineering
MAF v1 — DevUI for agents and workflows (Python)
Run MAF DevUI locally to test prompts, inspect tool calls and OTel spans, and register agents by code or directory discovery.
Deep DiveAI Engineering
MAF v1 — Production hardening (Python + .NET)
Add HTTP-layer production patterns for password reset, refresh-token rotation, reuse detection, and secret rotation in APIs.
Read article
Deep DiveAI Engineering
MAF v1 — Workflow Visualization (Python + .NET)
Render any workflow as Mermaid (GitHub-friendly) or Graphviz DOT (production runbooks). One line each, deterministic output, dark-mode-safe palette.
Deep DiveAI Engineering
MAF v1 — Declarative Workflows (Python + .NET)
Define a workflow in YAML, load it at runtime, and tweak orchestration without recompiling the Python or .NET app safely.
Deep DiveAI Engineering
MAF v1 — Human-in-the-Loop (Python + .NET)
Pause a workflow for human input, resume it with the same request_id, and keep the caller-side contract simple across runs.
Deep DiveAI Engineering
MAF v1 — State and Checkpoints (Python + .NET)
Persist workflow state at superstep boundaries, reload snapshots in a fresh process, and continue without losing progress.
Deep DiveAI Engineering
MAF v1 — Group Chat Orchestration (Python + .NET)
Use a centralized manager to pick speakers each round, compare round-robin and agent-driven strategies, and cap runaway loops.
Deep DiveAI Engineering
MAF v1 — Handoff Orchestration (Python + .NET)
Let a Triage agent route the conversation to Math or History specialists, handle handbacks, and guard against loops in production.
Deep DiveAI Engineering
MAF v1 — Magentic Orchestration (Python + .NET)
Use Magentic orchestration for autonomous planning: a manager keeps facts, delegates to workers, reassesses, and replans when stuck.
Deep DiveAI Engineering
MAF v1 — Agents in Workflows (Python + .NET)
Wrap a ChatClientAgent as a workflow executor, chain two translators, and hide adapter plumbing behind AgentWorkflowBuilder.