FDE Path

About this Path

This Path is an onboarding guide for engineers and product teams who build customer-facing solutions on top of D.Hub. It follows one end-to-end working flow: isolate a workspace, model the business domain, automate data, configure AI work, connect knowledge, and finish with a deployed experience.

The Path is organized as six lessons that follow the actual portal flow and screen sequence. Together, they give you one practical pass through how the main D.Hub resources fit together.

The complete Path forms one delivery flow from customer data to a deployed experience.

flowchart LR
    accTitle: FDE solution-delivery end-to-end flow
    accDescr: Prepare customer data in a collection, build ontology and pipelines, connect dashboards plus models and agents, add knowledge and tools, and finish with deployment and validation.
    collection[Collection and customer data] --> ontology[Ontology]
    collection --> pipeline[Pipeline]
    pipeline --> dashboard[Dashboard]
    ontology --> agent[Models and agents]
    knowledge[Knowledge, tools, connectors] --> agent
    agent --> deploy[Deploy and validate]
    dashboard --> deploy

Terms to know before you start

• Forward Deployed Engineer (FDE): An engineer who works in the customer environment to turn requirements into a working solution from data through user experience.
• Resource flow: The input-and-output chain connecting data, models, automation, AI, and presentation assets inside a collection.
• Large language model (LLM): The model that interprets natural-language requests and generates agent responses or tool calls.
• Retrieval-augmented generation (RAG): Searching relevant knowledge so a model can produce an evidence-grounded answer.

Prerequisites

• A D.Hub portal account with permission to create resources
• Finishing the Essentials Path first is recommended if you want a smoother start with the example flow

What you'll be able to do

• Understand how collections isolate workspaces and hold the main resources used in a solution
• Define entities and relationships in the Ontology Builder and shape a knowledge graph
• Build ETL pipelines that load data into datasets and ontology entities
• Turn datasets and ontology outputs into operational dashboards and widgets
• Register LLM models, configure agents, and connect tools and actors
• Attach data connections, knowledge bases, and RAG, then validate the deployed agent

What comes after this Path

If you want to continue with a domain-specific scenario, move on to a workshop:

• Retail Inventory Intelligence — Build a full scenario from pipeline setup to dashboard operations using realistic data

Check off each lesson as you finish it — progress is recorded automatically. Pick one up and start.