Most organizations approach "AI democratization" by giving coding assistants to engineers and simplistic chatbots to business users. Yet, building sophisticated, high-value AI agents remains a bottleneck restricted to a few specialists. In this talk we’ll explain how we enabled Forter’s engineers, analysts, and product managers to build their own autonomous agents. Within two weeks they’d built 80 agents. Some automated time-consuming things like customer onboarding, alert handling and support tickets, while others built chat agents expert in different parts of our system or ways of work.
We’ll talk about why this doesn’t have to be hard. What is the Minimum Viable Platform? What agent building environments should you support, and what are the trade-offs between them? Do you really need a full RAG solution? How did our in-house MCP server work, and why was it such a force multiplier? We’ll also cover practical tips on how to educate your future agent builders, manage token constraints and get your legal and security teams on side.
Interview:
What is the focus of your work these days?
I’m the Principal Engineer on Forter’s AI Engineering team. I have two main areas of focus: First, enabling employees with a wide range of technical backgrounds to build their own agents through training and platforms. Second, building agents to help our GTM teams work more effectively.
Who is your talk for?
Software Engineers and Architects who want to build the infrastructure that enables others to create.
Engineering Leaders and Product Managers looking for a proven cultural and technical playbook to accelerate AI adoption.
What is the motivation behind your presentation?
When we talk about “democratising AI” we are often very focused on providing engineers with coding agents, or providing less technical users with chat bots or simplistic agent builders. Building powerful agents end up being limited to a few engineers.
I believe that we can do much better. With the right focus we can get people across our organisations building their own high value agents. I want to show engineers some very achievable paths to enable exactly that, and on the way transform how their organisations view and use AI.
Speaker
Ben Maraney
Principal Engineer @Forter, Previously @BigPanda and @Klarna
I’ve been a Principal Engineer at Forter for the last 6 years. I lead our first LLM powered project in 2023, as well as our AI training for non-engineers. Previous focuses included ML and making data preparation easier for our data scientists, making data engineering accessible to our analysts and making data retention a topic that actually mattered to management. My talk will focus on my direct experience organizing our agentic sprint across our 250 engineer R&D team.
