Lessons
1Data wrangling under the gun47 min read
The forward-deployed job starts with a 250MB customer Excel, not a clean API. Streaming parse, type coercion, a canonical ontology, dedup/entity-resolution with dedupe, and a 20-sample quality eval — the flow that turns enterprise mess into something a model can stand on, fast.
- →Enterprise Data Quality
2APIs & integration patterns48 min read
Idempotency keys at the business-event level, full-jitter backoff that doesn’t synchronize into a thundering herd, HMAC-verified webhooks with a replay window and a dead-letter queue. The Stripe-grade integration envelope that stops your first deployment from double-charging the customer.
- →Reliable Integration Patterns
3The LLM app/agent layer49 min read
The agent loop is a few lines of glue: function calling → execute → append a tool message → loop. Structured outputs for anything that drives a write, RAG over the deduped data, guardrails as deterministic code, and an eval set written before the prompt. The AI layer that sits on top of your integrations without becoming a four-month rebuild.
- →Bounded AI Agent Layer
4Connecting to CRM/ticketing/doc stores48 min read
Three connector archetypes — CRM (Salesforce), ticketing (ServiceNow/Zendesk), doc stores (SharePoint/Confluence) — share one playbook: schema map, scoped per-customer OAuth, idempotent writes, verified webhooks for change-data-capture. Plus the integration-pattern choices, the SSO friction that eats weeks, and the per-document permission trap that leaks data.
- →Enterprise Connector Design
- →Reliable Integration Patterns
5Frontend for fast prototypes46 min read
A customer must click a URL and see their own data within 30 minutes of the meeting starting. The Vite + React + TS + TanStack stack, token streaming for perceived speed, zod-validated agent endpoints, an “evals as UI” page, and the demo-engineering discipline that keeps a credible prototype from leaking secrets or stalling the room.
- →Customer Prototype Delivery
6Capstone: an integrated assistant50 min read
Assemble the whole track into one engagement — deduped data, idempotent connectors, the agent layer, a streamed prototype — shipped on a customer’s clock with eval gates and guardrails. Then the full FDE interview loop: the coding round, the system-design round, and the client-facing behavioral round, scored on the five-dimension rubric.
- →Enterprise Connector Design
- →Bounded AI Agent Layer
- →Customer Prototype Delivery
Skills in this course
- 01Enterprise Data QualityProfile, normalize, reconcile, and monitor messy customer data.
- 02Reliable Integration PatternsChoose safe sync, webhook, queue, and idempotency patterns.
- 03Bounded AI Agent LayerWrap probabilistic model work with schemas, tools, guardrails, and evals.
- 04Enterprise Connector DesignConnect CRM, ticketing, identity, and document systems with isolation.
- 05Customer Prototype DeliveryBuild fast, secure, streaming interfaces that prove the customer workflow.