As generative AI moves from experimental novelties to mission-critical enterprise applications, the limitations of prompt engineering have become a significant bottleneck. While organizations often attempt to solve inconsistency by "hacking" prompts with more detail and complex constraints, these efforts inevitably suffer from drift and fail to deliver the repeatability required for operational reliability. This session argues that the industry must move away from treating prompts as configuration files and instead embrace a paradigm of architectural determinism.
We will explore how to transition AI agents from probabilistic "improvisers" into disciplined "routers" that execute deterministic artifacts. Attendees will learn a systematic framework for capturing and hardening success through:
- Human-in-the-Loop Governance: Evolving oversight from simple quality assurance into a catalyst for reliability and safe memory integration.
- Capturing Success Paths: Recording reasoning, identifying essential steps, and mapping technical dependencies into structured blueprints.
- Codifying Reasoning into Artifacts: Transforming validated workflows into executable code, documentation, and rules that eliminate hallucinations and randomness.
- Building the Agent Skill Store: Creating a centralized library of proven solutions that allow agents to solve recurring problems with procedural precision rather than speculative generation.
By shifting the focus from linguistic hacks to system architecture, technology leaders can build GenAI systems that are truly scalable, auditable, and trustworthy.
Speaker
Hugo Guerrero
Principal Developer Advocate @Kong
For more than twenty years, Hugo Guerrero has helped organizations transform ideas into scalable and reliable software, working as a Developer, Architect, and Software Development Manager across a range of industries, including financial services, logistics, and transportation.
In his current role, Hugo Guerrero focuses on APIs, AI-driven capabilities, and event-based architectures, supporting teams in modernizing how they build, integrate, and operate software. He collaborates closely with engineering and product teams to guide adoption, enhance developer experience, and shape the technical narrative around emerging architectural patterns.
Hugo Guerrero’s mission is to serve as a technical educator and a bridge for the broader community, translating complex concepts into clear and practical insights that accelerate understanding and adoption. This includes leading webinars, workshops, and conference sessions, as well as developing enablement materials that support both partners and internal teams.
He is driven by a longstanding passion for software development and a strong commitment to continuous learning.
Session Sponsored By
Kong Inc., a leading developer of API and AI connectivity technologies, is building the connectivity layer for AI. Trusted by the Fortune 500 and AI-native startups alike, Kong’s unified API and AI platform enables organizations to secure, manage, accelerate, govern, and monetize the flow of intelligence across APIs and AI traffic — on any model, any cloud.
Video is not available
