AI Fluency & Building · Internship Role

AGENTIC AI DEVELOPER

Build AI agents for research and reporting tasks, with tools, memory and guardrails, and with humans in control.

SKILLS YOU'LL WALK AWAY WITH

PythonTool use / function callingAgent design patternsAgent evaluationGuardrails

PROOF-OF-WORK TASKS

8 TASKS

Start at Beginner and work up. Every task produces something real you can show: a report, a tool, a guide or a demo. Publish it on LinkedIn or GitHub as your proof of work.

Beginner

LEVEL 01

01Build a simple tool-using script where an LLM can call one function (for example, read a local file). Explain each step of the tool-use loop in your README.
02Build a 'news digest' agent that reads 5 cybersecurity RSS feeds and writes a daily 5-bullet summary with source links.
03Write a short design doc comparing workflows and agents: when should a fixed pipeline beat a free-running agent? Use 3 examples.

Intermediate

LEVEL 02

01Build a research agent that takes a company name, searches public sources, and writes a 1-page security posture brief with citations. It must flag anything it could not verify.
02Add guardrails to your agent: a step limit, an allowed-tool list, a cost cap, and a human approval step before any external action.
03Create a 10-question evaluation set with expected answers, and measure your agent's accuracy, cost and time per task.

Advanced

LEVEL 03

01Build a multi-step reporting agent (planner → researcher → writer → reviewer) that produces a weekly threat report. Log every tool call, and make the reviewer catch unsupported claims.
02Red-team your own agent: try prompt injection through web pages it reads, document what broke, and ship fixes with before/after tests.

HAND-PICKED FREE RESOURCES

6 RESOURCES

FREE COURSE