IntermediateAI Agents with LLMs project · about 12 hours

Helpdesk agent with memory and human hand-off

Build a helpdesk agent that resolves simple requests, remembers context and hands others to humans.

The real-world brief

Level-1 tickets fill the queue and the company wants routine ones handled automatically.

What you will build

  • A workflow for password reset and access requests
  • Conversation memory
  • An approval step before actions
  • Hand-off to a human with a summary

Skills you will practise

Agent workflowsState and memoryApprovalsHand-off design

Tools

PythonLangGraph or a similar frameworkAn LLM APIA ticket system API or mock

Project tasks

The detailed steps, checklists and trainer review open in the SOTT student portal for enrolled students.

1Workflow Steps in portal
2Memory Steps in portal
3Approval Steps in portal
4Hand-off Steps in portal

How companies use AI Agents with LLMs

Insurance: Policy questions answered in seconds. Agents searched long policy PDFs to answer customer questions. A RAG assistant searches the policy library and answers with citations to the exact clause. Average handling time fell and answers became more consistent.

IT services: Helpdesk agent. Level-1 tickets like password resets and access requests filled the queue. An AI agent with approved tools resolves routine tickets and hands others to humans with a summary. Routine tickets are resolved faster and engineers focus on harder issues.

Legal: Contract review. Lawyers spent hours finding clauses in contracts. An LLM pipeline extracts key clauses and risks into a structured summary for review. First-pass review time dropped sharply, with lawyers checking every output.

Questions

Can I add the Helpdesk agent with memory and human hand-off project to my resume?

Yes. Once your trainer approves it, list it under Projects with the tools you used, one line on the problem it solves and a link to your repository and demo video.

How long does this project take?

About 12 hours of hands-on work, and it fits best after week 6 of the AI Agents with LLMs course.

Where do I get the step-by-step tasks?

Enrolled students get every task with steps, "done when" checks, the submission form and a trainer review in the SOTT student portal.