Travis Gosselin

Travis Gosselin
   

Distinguished Software Engineer, SPS Commerce

Travis is a Distinguished Software Engineer at SPS Commerce, the world's largest retail network, where he leads innovation, drives meaningful change, and amplifies team productivity. With a passion for technology and a talent for simplifying complex architectures, Travis focuses on creating seamless Developer Experiences and fostering high-performing teams. A seasoned speaker, architect, and blogger, he thrives on sharing insights, exploring emerging technologies, and championing impactful development practices.

Presentations

A Developer’s Guide to Building Your First MCP Server

5:00 PM MDT

As AI agents become a first-class part of software systems, learning to build MCP servers is becoming as fundamental as learning to build RESTful APIs. Many engineers already interact with MCP through IDEs and agent tooling, and building these servers introduces a new set of design considerations beyond traditional request/response APIs. While MCP servers share familiar API concepts, the move to agent-driven, probabilistic clients changes how engineers think about contracts, tool design, output shaping, state management, error handling, and spec adoption. Building MCP servers is emerging as an important capability for all developers and teams.

In this session, we’ll start with what MCP is and how the protocol defines resources, tools, prompts, and elicitations. We’ll walk through the practical decisions involved in building your first MCP server, including transport choices, authentication, tool and resource granularity, server instructions, and context engineering. You’ll see how MCP aligns with and diverges from traditional API design, explore framework options, preview generated MCP servers, and learn how to design for scaling, versioning, gateways, and deterministic error handling in real-world systems.

Crafting Consistent APIs at Scale: Embracing Linting & Reusable Models

1:30 PM MDT

In an era where digital transformation and AI adoption are accelerating across every industry, the need for consistent, scalable, and robust APIs has never been more critical. AI-powered tools—whether generating code, creating documentation, or integrating services—rely heavily on clean, well-structured API specifications to function effectively. As teams grow and the number of APIs multiplies, maintaining design consistency becomes a foundational requirement not just for human developers, but also for enabling reliable, intelligent automation. This session explores how linting and reusable models can help teams meet that challenge at scale.

We will explore API linting using the open-source Spectral project to enable teams to identify and rectify inconsistencies during design. In tandem, we will navigate the need for reusable models—recognizing that the best specification is the one you don’t have to write or lint at all! These two approaches not only facilitate the smooth integration of services but also foster collaboration across teams by providing a shared, consistent foundation.

Engineering the Context That Powers Developers and Agents

3:15 PM MDT

Engineering teams have always struggled with fragmented documentation, tribal knowledge, and inconsistent standards. What we now call context engineering is closely related to long-standing knowledge management challenges, but the impact is amplified as AI agents enter everyday development workflows. Gaps, conflicts, and poorly organized context slow developers and lead automated systems to produce unreliable results. Without deliberate strategies, organizations risk scaling confusion instead of capability.

In this session, we will explore practical ways to engineer context and knowledge management that works for both developers and AI agents. Using real examples from building centralized, docs-as-code platforms, we will cover how to organize internal documentation, define ownership between team and platform context, and validate content for gaps and conflicts. We will also examine shared context engineering strategies such as summarization, retrieval, and externalization, including how MCP-based approaches improve clarity, scalability, and execution for humans and machines alike.

AI Enablement in the SDLC: A Strategy Sampler

5:00 PM MDT

AI enablement isn’t buying Copilot and calling it done; it’s a system upgrade for the entire SDLC. Code completion helps, but the real bottlenecks live in reviews, testing, releases, documentation, governance, and knowledge flow. Achieving meaningful impact requires an operating model: guardrails, workflows, metrics, and change management; not a single tool.

This session shares SPS Commerce’s field notes: stories, failures, and working theories from enabling AI across teams. You’ll get a sampler of adaptable patterns and anti-patterns spanning productivity, systems integration, guardrails, golden repositories, capturing tribal knowledge, API design, platform engineering, and internal developer portals. Come for practical menus you can pilot next week, and stay to compare strategies with peers.