A conference can leave you with a full notebook and a Monday morning that looks exactly like the previous one. You watched demonstrations, discovered tools and met people. Then work resumed, and everything waited in a browser tab.
DevFest 2026 is an opportunity to make the return more useful. In its October 30 announcement, Google places the community events between October 1 and December 31, with workshops and tracks covering AI, cloud, Android and web. That global calendar does not establish a date in a Moroccan city: check the local organiser’s programme.
Choose a question before choosing sessions
For a small team, a day away needs an objective specific enough to survive the return to work. “Understand AI agents” is a starting point. “Find out whether we can automate ticket classification without exposing customer data” gives you a question to take into a workshop.
This fictional scenario shows how to turn a broad theme into a recognisable problem. The approach also works for a slow Android app or a cloud deployment that is hard to reproduce: bring a situation, known constraints and the decision you want to make.
Then read the prerequisites. A workshop requiring an installed environment offers something different from an introductory talk. Confirm level, language, accounts and any required credits on the organiser’s page. The global announcement does not establish an individual event’s admission or workshop conditions.
A demonstration starts the conversation
Once a demo works, ask what remains outside its frame. Which dataset does it use? What permissions were granted? What happens when a service responds incorrectly or stops responding? Can you export the result and reproduce the exercise elsewhere?
These questions help turn enthusiasm into something usable. For a team serving clients in Morocco, content languages, existing tools and connection conditions can change a solution’s value. Test those conditions yourself rather than inferring a fit from an example on stage.
Our reading of the Devoxx Morocco 2026 programme covers another meeting point for cloud and AI engineering. It provides context, not a claim that DevFest has the same programme, format or conditions.
Bring back an experiment small enough to finish
A useful return can start with a brief agreement between the attendee and their team. This table proposes a working method, not a DevFest programme requirement.
| When | What to prepare or bring back | Why it helps |
|---|---|---|
| Before | A specific question and an example without sensitive data. | Choose sessions that inform a decision. |
| During | A reproducible example, dependencies and limitations. | Separate what you watched from what you can repeat. |
| After | An owner, a short deadline and a stopping criterion. | Prevent an experiment drifting without an answer. |
In the ticket example, a trial could use a small set of anonymised or synthetic requests that the team has already classified. Compare mistakes and correction time with the existing method, while keeping reply sending outside the experiment. This is a proposed test, not a measured result or a guarantee of anonymisation.
Decide what would make you drop the idea: excessive integration work, fragile results or savings that disappear after review. An experiment that leads you not to adopt a tool can be valuable. It prevents a more expensive decision.
Find events through the DevFest directory linked by Google, then confirm dates, location and conditions with the organising GDG. To evaluate the day, look at the work it enabled. Session counts show that you were busy. The resulting decision says more about whether attending was worthwhile.




Discussion
Add your perspective.
A question, a different perspective or experience to share? Add to the conversation.