Effective practices with coding agents
Where an agent genuinely saves time and where it costs you more than it saves. How to scope a task so the result is reviewable, and what to do when the agent confidently goes wrong.

Where an agent genuinely saves time and where it costs you more than it saves. How to scope a task so the result is reviewable, and what to do when the agent confidently goes wrong.
Agents are only as good as what they can see. Structuring a codebase, a prompt and a working set so the model has the right context — and stops inventing the parts it lacks.
Using the model as a reviewer rather than an author: finding real defects, pushing back on plausible-looking output, and telling the difference between a confident answer and a correct one.
Writing the spec first so the agent builds the right thing. Turning a spec into a plan, a plan into tasks, and keeping all three honest as the work changes underneath them.
Developers and students who want to use coding agents well, and get output they can trust and ship.
Yes. The session is taught live, and registered attendees get the recording afterwards.
Want this workshop for your class or team? Talk to us →