Y

AI · Skills roadmap

Forward Deployed Engineer projects

Build AI systems inside customers' businesses and make them work in the real world.

  • 40 to 64 weeks to job-ready, part time
  • 49 lessons across 20 skills
  • 12 portfolio projects

61 lessons and projects. Progress saves in this browser.

Start the roadmap

Beginner

4 projects
  1. 1Communication · Customer-facing delivery · SQLFrom a vague ask to a pilot plan8 h · 1 dataset
  2. 2LLMs and NLP · Machine learning · PythonRoute banking questions to 77 intents10 h · 1 dataset
  3. 3LLMs and NLP · Prompt engineering · PythonA support reply assistant with structured output10 h · 1 dataset
  4. 4Git and GitHub · REST APIs and HTTP · TypeScript and JavaScriptSync support tickets into a helpdesk API with TypeScript12 h · 1 dataset

Intermediate

4 projects
  1. 1LLMs and NLP · LLM evaluation · PythonA cited question-answering assistant over the Kubernetes docs20 h · 1 dataset
  2. 2CI/CD · LLMs and NLP · LLM evaluationAn evaluation harness that blocks bad prompt changes16 h · 1 dataset
  3. 3AI agents and tool use · Python · Model Context Protocol (MCP)An MCP server for World Bank development data14 h · 1 dataset
  4. 4LLMs and NLP · Monitoring and observability · Docker and containersAn LLM triage integration driven by helpdesk webhooks20 h · 1 dataset

Advanced

4 projects
  1. 1AI agents and tool use · LLM evaluation · Monitoring and observabilityA data question agent with guardrails and an eval suite28 h · 1 dataset
  2. 2Customer-facing delivery · LLMs and NLP · LLM evaluationRun a customer pilot end to end32 h · 1 dataset
  3. 3Cloud platforms (AWS, Azure, Google Cloud) · Security compliance and data protection · Customer-facing deliveryPass a customer's security review with an LLM deployment24 h · 1 dataset
  4. 4Deep learning · LLMs and NLP · LLM evaluationFine-tuning versus prompting, decided with numbers24 h · 1 dataset