AI Agents & Workflows
Fictional sample profileMaya teaches agent design as workflow engineering: well-scoped tools, explicit state, clear stopping rules and a plan for every way a step can fail.
How Maya reviews work
Maya reviews agent projects by reading run traces. Feedback focuses on tool boundaries, state handling and whether the agent stops, retries or escalates at the right moments.
01
Maya's modules start from a simple position: an agent is a loop with a model inside it, and the loop deserves as much design attention as the prompt. Learners define narrow, typed tools, persist workflow state and set budgets for steps, time and cost.
She covers when a fixed workflow is a better choice than an open-ended agent, how to add human approval for actions with side effects, and how to make runs traceable so failures can be reproduced. Guardrails and permission boundaries are treated as core features, not add-ons.
Maya encourages learners to test their tools without a model first, and to keep a log of failed runs as a source of new test scenarios.
This is demo content. It does not describe a real person, employer, qualification or track record.
02
03
plan
select tool
validate args
execute
observe
AgentsGuide10 min read
An agent is a loop that lets a model choose actions. This guide covers how to bound that loop, constrain tools, validate arguments, and design an explicit path for when things go wrong.
By Maya Chen · sample author
output · TicketTriage
{
"category": "billing",
"priority": "high",
"order_id": "48213",
"needs_human": true
}
FundamentalsGuide8 min read
When an LLM's output feeds other code, treat it like any untrusted input: define a schema, validate strictly, retry with specific feedback, and fail loudly when the model can't comply.
By Alex Morgan · sample author
Build the foundations, projects, and habits needed to design more dependable AI applications.