Agentic AI in the Wild - Callstack & Qodo Dev Community & ReactVision & Codex Community in Boston
Codex Community & Qodo Dev Community & ReactVision & Callstack are coming to Boston!
AI agents are moving into production, and the hard part isn't getting one to work — it's making it work reliably, day after day, on real codebases. This meetup gathers builders, founders, and practitioners who are past the demo stage and deep in the mess of integrating agents into CI/CD pipelines, developer platforms, and knowledge systems.
We've lined up practitioners who'll share what's actually working (and what isn't) in their own agentic systems, followed by time to talk it through with the people sitting next to you.
Four talks, each from a different point in the agent workflow. You'll see what you can catch locally — mutation testing, coverage-guided fuzzing, CPU-level profiling on your laptop's efficiency cores — before you start paying per run on a real device cloud. You'll watch an AI agent build a working AR scene live on stage through an MCP server, no Unity, no new engine, just the React Native stack you already know. You'll get a practical method for encoding your team's standards into machine-readable guardrails so AI-generated code respects your architecture instead of quietly eroding it. And you'll hear how a coding agent, re-prompted on a loop every 30 minutes, chewed through a Vite-to-Expo migration of a large React app one manageable piece at a time — including where it broke. Then a panel to argue about all of it.
Agenda highlights:
6:00 PM — Doors Open & Pizza & beers
6:30 PM — Intro words
6:40 PM — Fuzz First, Farm Later: Local Validation Before Costly Cloud / Matt Hargett (Callstack)
7:00 PM — How to Embed Code Quality into SDLC / Nnenna Ndukwe (Qodo)
7:15 - 7:55 PM Panel discussion
8:00 - 8:15 PM - Prompt to Portal: How AI Makes Spatial Computing Accessible / Oliver Edis (Reactvision)
8:15 - 8:30 PM — Large Codebase Migration with Loop Engineering / Rafael Mendiola (okthink)
9:00 PM — Closing
Talks abstracts:
Fuzz First, Farm Later: Local Validation Before Costly Cloud / Matt Hargett (Callstack)
Agent-driven testing on real device clouds closes an important part of the verification loop — and the bill scales with your commit rate. If you're revalidating on every merge, that curve gets steep fast. This talk covers what you can catch before you touch a device: branch mutation testing, coverage-guided fuzzing, and deep CPU-level profiling. Pinning L1 misses, prefetch stalls, branch mispredictions — run on your laptop's efficiency cores as a stand-in for constrained phone, watch, XR, or TV silicon. I'll walk through a case study where I put agents on all three ahead of an AI-driven rearchitecture away from React.
2. How to Embed Code Quality into Your ADLC with Codex / Nnenna Ndukwe (Qodo)
AI is a force multiplier that turns weak standards into architectural chaos. As code review becomes the ultimate bottleneck, engineering teams must bridge the gap between human intuition and machine output. In this talk, we'll explore how to codify development best practices into machine-readable guardrails. We'll walk through how to leverage context engineering at scale to ensure your AI code tools respect your system’s design, preserving long-term maintainability without sacrificing the speed of the AI era.
You'll learn how to embed code quality into the SDLC from AI-generated output with Codex and multiple AI dev tools. They will leave with a practical methodology for encoding best practices into their workflow, where developers already operate (CLI, IDE, and Git).
3. Prompt to Portal: How AI Makes Spatial Computing Accessible / Oliver Edis (Reactvision)
AI is opening up spatial computing, and pairing it with React Native makes it more accessible than it's ever been. I'll show exactly what that looks like prompting an AI agent live on stage to build a working AR scene through ReactVision's MCP server, no Unity, no new engine, just the React Native stack you already know. We'll talk through what the agent actually has access to under the hood, what it gets right, what it doesn't yet, and why building for physical space is a different kind of challenge than building for a screen.
4. Large Codebase Migration with Loop Engineering / Rafael Mendiola (okthink)
Moving a large React app from Vite to Expo is not a simple framework upgrade. Hundreds of components, styles, browser APIs, and tests must be converted without breaking the product. Traditionally, we would create a huge project plan, divide it into dozens of tickets, and spend weeks coordinating and reviewing the work. Instead, we created an Agent Loop: a coding agent repeatedly received the same prompt every 30 minutes, found one manageable part of the migration, converted it, ran tests, recorded its progress, and left the code ready for the next run.
This talk explains how that loop turned an overwhelming migration into a series of small, repeatable steps. You’ll learn how to give agents clear rules, help them choose the next task, verify that their changes are correct, and prevent them from repeating work or drifting away from the goal. You’ll also see where the approach failed, why strong tests and feedback loops mattered more than detailed ticket planning, and how your role changes from writing every line yourself to building a system where agents can make safe, steady progress.