Building Coding Agents with Tool Execution
by DeepLearning.AI
Build agents that write and run their own code safely in sandboxed cloud environments.
Overview
Building Coding Agents with Tool Execution is a 9-lesson (~1 hour 21 minute) short course from DeepLearning.AI in collaboration with E2B, taught by E2B's Tereza Tizkova and ML engineer Francesco Zuppichini. It explains what a coding agent is and how it differs from a tool-calling agent: instead of choosing from predefined functions, the LLM reasons step by step, writes and executes code, reads the output, and plans its next action until it reaches a final answer. The course centers on safe code execution — comparing local execution, containers, and secure sandboxed microVMs, and running agent-generated code in isolated E2B cloud sandboxes so untrusted code can't touch your systems. Learners build a data-analyst agent that explores CSV files with pandas and produces visualizations, and finish by building a full-stack coding agent that edits multiple files and generates a working Next.js web app entirely inside a sandbox. Includes 5 code examples and a graded assignment.
At a Glance
- Topic
- Agentic
- Level
- Intermediate
- Format
- Course
- Cost
- Freemium
- Duration
- ~1 hr 21 min, 9 video lessons, self-paced
- Provider
- DeepLearning.AI
- Hands-on
- Yes — code/exercises
- Certificate
- None
What You’ll Learn
- ✓How a code-writing-and-executing agent differs from a fixed tool-calling agent
- ✓Building an agent loop that writes Python, runs it, reads results, and iterates
- ✓Trade-offs between local, container, and microVM sandbox execution environments
- ✓Running untrusted agent-generated code safely in isolated E2B cloud sandboxes
- ✓Building a data-analyst agent (pandas + visualizations) and a full-stack agent that generates a Next.js app
Highlights
- •Taught by the E2B team, who build the code-execution sandbox infrastructure
- •Focuses on the security model of running LLM-generated code, not just prompting
- •Two full projects: a data-analyst agent and a full-stack Next.js app builder
Who It’s For
Best For
- ✓AI engineers building code-execution or data-analysis agents
- ✓Developers who need to run LLM-generated code safely in production
- ✓Teams evaluating sandboxing options (containers vs. microVMs) for agents
Prerequisites
- •Intermediate Python
- •Familiarity with LLM APIs and basic agent concepts
- •Helpful: comfort with the command line and Docker-style containers
FAQ
What is Building Coding Agents with Tool Execution?
A hands-on short course from DeepLearning.AI and E2B on building coding agents that go beyond fixed function calls — agents that write Python, execute it, read results, and iterate. It teaches AI engineers how to run untrusted agent-generated code safely across local, containerized, and microVM sandbox environments.
Is Building Coding Agents with Tool Execution free?
Building Coding Agents with Tool Execution offers free content, with paid options for certificates or premium features.
What level is Building Coding Agents with Tool Execution for?
Building Coding Agents with Tool Execution is aimed at a intermediate audience. Recommended background: Intermediate Python, Familiarity with LLM APIs and basic agent concepts, Helpful: comfort with the command line and Docker-style containers.
How long does Building Coding Agents with Tool Execution take?
Expect roughly ~1 hr 21 min, 9 video lessons, self-paced. Most learners work through it at their own pace.
What will I learn from Building Coding Agents with Tool Execution?
You'll learn: How a code-writing-and-executing agent differs from a fixed tool-calling agent; Building an agent loop that writes Python, runs it, reads results, and iterates; Trade-offs between local, container, and microVM sandbox execution environments; Running untrusted agent-generated code safely in isolated E2B cloud sandboxes; Building a data-analyst agent (pandas + visualizations) and a full-stack agent that generates a Next.js app.