The Context Engineering Required For a Killer Marketing Agent

Most AI marketing agents underperform. They produce generic copy, ignore brand voice, and require more hand-holding than just doing it manually. The problem usually isn't the model. It's the context.

Context engineering is the discipline of designing what your AI agent knows, how it knows it, and when it can access it. Done well, it's the difference between an agent that hallucinates and one that ships quality, on-brand marketing work with minimal oversight.

In this session, Adam K. Stinson, Founder of Dark Horse Growth, walks through the full picture of what it takes to build a marketing agent that actually works. He'll cover:

• Marketing principles for agents — the fundamentals of campaign design you need to know before you can expect an agent to do valuable work for you.
• Architecture of an autonomous marketing agent — the components that make real autonomy possible: tools, memory, decision loops, and orchestration
• The crucial part: context — what your agent needs to know (brand positioning, ICP profiles, competitive intelligence, content history, voice guidelines), how to structure it, and how to test whether it's actually working
• Putting it all together — a synthesis framework for building, validating, and iterating on your marketing agent

This is a practitioner session designed for builders and operators. Adam will share frameworks and real examples from his consulting work, and the Q&A will be open for your specific use cases.