AI agent development
An agent that handles a few hand-picked examples is quick to build. One that uses company systems, stays within its permissions and holds up on real requests needs a good deal more engineering.
This hands-on training is for software and engineering teams that will build and run AI agents themselves. It covers the whole path, from choosing a task to running the agent in production.
What the training is about
Most of the work in an agent sits outside the model. The task has to be clearly defined, the tools kept narrow and the permissions and approval steps set. The agent then needs testing on real cases and monitoring once it is live. Whether it can be trusted with real work depends on these decisions, whichever framework or model you use.
Participants build an agent step by step on a realistic task, ideally one from their own work if internal rules allow it.
Who it is for
Software developers and backend teams
Building agents that call company APIs, databases and services, and writing the tools they use.
Platform and DevOps teams
Providing the environment, credentials, logging and monitoring that agents need in production.
Technical leads and architects
Deciding where an agent fits, which permissions it gets and which standards agent work follows across teams.
Product engineers
Owning an agent feature in a product and keeping track of how it behaves with real users.
Topics
We agree with you in advance how much time each topic gets, based on your team's work.
- Choosing a task that suits an agent, defining it clearly and recognizing when ordinary automation is enough
- Turning APIs, databases and documents into narrow tools, directly or through MCP servers
- Permissions and approval steps, set separately for each action
- Working with company data, and treating text in emails, documents and tool outputs as untrusted
- Testing agents on real cases, edge cases and tool failures with a repeatable evaluation suite
- Cost per task, latency and model choice for each step
- Running agents in production: logging, monitoring and new test cases from real use
Frequently asked questions
Which frameworks and models are used in AI agent development training?
We work with the languages, frameworks and models your team already uses. If you have not settled on any yet, we choose them with you before the training starts.
What do developers need to know before AI agent training?
Programming experience and some familiarity with APIs are enough. No machine learning background is needed.
Can we build an agent for our own use case during the training?
Yes. If your internal rules allow it, we build the exercises around a task you plan to give an agent. The team leaves with a clear view of what that agent needs before it can go into production.
Plan a training program with us
Tell us about your team and what you want them to be able to do. We'll get back to you.
From our blog
All articles
How much does an AI agent cost? It depends on what you let it do
AI agent projects can cost $8,000 to $250,000 or more. Estimate how permissions, integrations, exceptions, data readiness, and oversight affect the budget.
Read article
How to build an AI chatbot with your own data
Connecting documents to a model is the easy part. Learn what a company chatbot needs: source rules, system connections, access control, handover, and testing.
Read article
You're not doing AI. You're doing integration.
Most AI projects depend on integration. Company data, business rules, human review, and workflow design turn a capable model into a useful product.
Read article