Commit Your Code 2026
2026-09-02..2026-09-04
-- online --
[Commit Your Code 2024, Commit Your Code 2025, Commit Your Code 2026]
as the organisers put it
The archived 2026 Commit Your Code Conference: six technical tracks, 100+ speakers, and two days at the Capital One Campus in Plano, Texas, September 3–4, 2026.
who it's for
platform engineers, ai engineers, engineering managers, site reliability engineers, frontend developers, security engineers, developer advocates
topics
agent architecture and orchestration; Agent2Agent protocol implementations; MCP servers and connectors; Claude skill development; agent-to-ui secure communication; model governance and AI BOMs; memory integrations for agents; automated agent evaluation and scoring; observability and profiling for agents; React Server Components and Suspense; AST-based frontend automation; serverless durable executions
| title | author | length |
|---|---|---|
| Keynote | Kevin J. Scott | 49m |
Organization standardizes on Cursor as the single agent ecosystem, centralizes skills and configs in one hub repo served over MCP, and replaces local dev… | ||
| The Put That Sh*t On Agent: Building Multi-Agent AI Systems That Actually Talk to Each Other* | William Hill | 28m |
A working multi‑agent wardrobe app connects a central orchestrator to specialist agents via the Agent2Agent (A2A) protocol, uses MCP to fetch live weather and… | ||
| Claude Skills: Automate the Boring Stuff | Jacob Orshalick | 25m |
Demonstration builds reusable Claude skills end to end: a golf-hotel finder that calls an Expedia MCP connector, geocodes the course, filters by distance… | ||
| Building AI Agents in TypeScript | Raja Krishna | 23m |
Shows how to build production-ready AI agents with Mastra and TypeScript by wiring agents to real tools (example: get budget versus actual tool), designing… | ||
| Sociotechnical Platform Engineering | Chris Corriere | 25m |
Zoomed-out framework linking DevOps and SRE across a platform value stream, with concrete artifacts: a reusable repo that boots a markdown blog as a smoke… | ||
| React - Server Components, Suspense, and the new mental model | David Kea | 22m |
A demo restaurant app called Haleyena Poke and Grinds models a $10,000/month small business competing with 50+ local options and shows how React Server… | ||
| Agents are not Apps: Stop building interfaces, Start building coworkers | Boston Cartwright | 25m |
An agent named Ella runs for five months over Telegram, handles RSVPs, builds an RSVP website, creates a budget tracker, and books four vendors for a wedding… | ||
| AI-Driven Supply Chain Security: Using AI-BOM to Detect Hidden Vulnerabilities | Anitha Dakamarri | 25m |
Claims that an AI-BOM engine supplements SBOM by cataloging models, training data, prompts, agents, APIs and cloud resources, then correlates them into an AI… | ||
| AI Governance in Financial Services: What Every Developer Should Know | Belle Duran | 24m |
Explains that the model is one component and everything wrapped around it — identity permissions logs and clear workflows — decides deployability, not accuracy… | ||
| Remote Agents, or: How I Fixed Our Page's Performance Without Really Doing Anything | Eric Karschner | 21m |
Built an agent pipeline that pulls DataDog RUM browser profiler telemetry, finds hot functions via profile aggregation, spins up a local dev server to… | ||
| The Death & Resurrection of the IDE: Why Agents Need Semantic Understanding | Jason Torres | 13m |
Argues that IDEs do not die but evolve into a supervision cockpit that gives agents semantic sight, showing code as AST/dataflow graphs rather than strings… | ||
| The Series Finale Problem: What Happens When Your Brand Makes a Promise You Can't Keep | Jeremy Barger | 23m |
Explains the Series Finale Problem with specifics: How I Met Your Mother hides the titular character for 32 minutes (0.007%) of its 4,560 minutes, Apple Maps… | ||
| CSS as a Game Engine: Building Playful, Stateful Web Experiences Without JS | Aaron Krauss | 27m |
Argues that HTML+CSS alone can act as a game engine by treating UI controls as state, using radio buttons as FSM and checkboxes for inventory, CSS counters for… | ||
| Good vibes, smooth payments: Tailoring Stripe to any UI | Anna Spysz | 23m |
Claim: you can make Stripe look like it always belonged on any site even a GeoCities throwback. Demonstrations use Payment Links (hosted pages, minimal code)… | ||
| Engineering Real-Time Data Synchronization Across Multi-Cloud Systems | Eunice Kamau | 24m |
Describes a three-layer architecture that separates data acquisition, a synchronization orchestrator, and application access, with event-driven propagation… | ||
| From Requirements to Reality: How Solutions Architects Think, Build & Decide | Raviteja Sankuratri | 25m |
Explains concrete architect moves: uses Deploy on AWS plugin to generate IaC and an AI-driven diagrammer, compares Lambda vs EC2 with cold-start, concurrency… | ||
| Build Failed Startups—Not Side Projects | Ramiro Beltran | 25m |
Argues that choosing intent to go to production changes every decision: code is deployable, flows navigable, and gold‑plating stops. Recommends shipping 0.0.1… | ||
| How to Use Scaffolding to Make Your Coding Agents Less Dumb | Quincy Larson | 20m |
Scaffolding means predefining folder structure, dependencies, design system, tests and deployment so agents make predictable choices instead of inventing extra… | ||
| The Devil Was Never in the Details: Code, AI, and the Rise of the Deep Generalist | Tim Banks | 25m |
Argues that the seams between services—not individual platforms—determine success, and that AI magnifies the need for cross-domain practitioners who stitch… | ||
| How to Test AI Features Before Your Users Do | Ron Lyle Dagdag | 24m |
A lightweight eval loop is built around an anonymized knowledge assistant in Microsoft Foundry that catches three concrete failures: unsupported claims… | ||
| If You’re Not Shipping, You’re Not a Leader | Zackary Chapple | 25m |
Concrete examples: moved a JBoss‑based DurEcoEmos server off the factory floor to avoid $16,000/minute downtime; built a Node + Cheerio DOM‑scrape backend that… | ||
| Writing a Container Runtime in Go | Michael Duren | 22m |
Builds a minimal runtime in Go that uses Linux syscalls directly: clone flags for namespaces (PID, mount, UTS), chroot to swap the process root, explicit… | ||
| Building Stelvio: An Open Source Journey to Deploy Python to AWS | Bas Steins | 19m |
Stelvio is an open-source, Python-only framework that maps application components (REST API, DynamoDB, Lambda) to AWS and generates automatic IAM and… | ||
| Fixing Potholes And Broken Roads: Devs Can Troubleshoot the Network With Ops Too! | Marino Wijay | 23m |
Local container fails because the app binds to localhost instead of the host network so changing localhost to 0.0.0.0 fixes the curl connection; Kubernetes… | ||
| The Evolution of Divot: How Agents Earn Their Autonomy | Michael Gyarmathy | 25m |
Divot automates three concrete workflows and shows measurable wins: a feature-flag pipeline that cleans stale flags after 15 days by dispatching coding agents… | ||
| 7 Leadership lessons for the engineer that AI can't replace | Nachiket Desai | 15m |
Seven concrete leadership lessons for engineers who use AI: own the outcome not the code, think before you prompt, translate technical trade-offs into business… | ||
| Baseline Magic: The Art of Intent-Driven CSS | Linda Thompson | 25m |
Live demo builds a small apothecary dashboard called Sage and Sundry using modern CSS: custom properties with @property for animatable colors, cascade layers… | ||
| Squeezing Java Containers: Shrinking Your Kubernetes Footprint with jlink | Mark Mackenzie | 20m |
Shows a concrete refactor path from a 523 MB Tomcat WAR to a 121 MB Alpine image using jlink custom JRE, cutting image size ~77% with zero business logic… | ||
| Thinking Architecturally | Nathaniel Schutta | 25m |
Explains how to weigh trade-offs with concrete rules: require a business problem and measurable payoff before adopting tech, avoid hoisting Kafka for 40,000… | ||
| AI for Java Developers | Dan Vega | 25m |
Java runs inference-grade AI in production and two thirds of AI compute is now inference with 55% of cloud spend on inference and 74% of builders saying most… | ||
| From APIs to Agents: Enterprise AI using Java, Ollama & MCP | Prathap Raghavan | 26m |
A Java Spring Boot agent orchestrates Ollama LLM via an MCP client/server, exposing five concrete tools — create, read, update, delete, list — backed by a JSON… | ||
| 5 Mistakes Developers Make When Building Java APIs (and How to Fix Them) | Yodit Weldegeorgise | 21m |
Five concrete API mistakes with Java and Spring Boot come with specific fixes and a demo repo: inconsistent naming (use kebab-case URLs), misplaced validation… | ||
| OOPs!... I Did it Again: When Good Objects Go Toxic | Trey Clark | 25m |
Shows Java-based library examples that break a bloated Title/Item parent into domain models, moves catalog/copyright/metadata responsibilities into separate… | ||
| Prickly by Nature: TanStack in the Wild | Sara Baqla | 24m |
A data-heavy web app called IMRS Biodiversity Explorer aggregates 1,200 species, five years of station weather, 90 named places, and a live iNaturalist feed… | ||
| Utilizing the JavaScript Abstract Syntax Tree (AST) for Automated Frontend Lifecycle Management | Tim Damen | 25m |
Organization builds a CLI scaffold and a migration pipeline using NX AST utilities to parse JavaScript, run deterministic codemods, and unit test them with the… | ||
| Architecting AI in Angular: WebMCP, A2UI, and the Agent Experience | Ankita Sood | 25m |
Walks through wiring WebMCP into Angular (navigator.modelContext) to register callable tools so agents don’t screenshot the DOM, shows using signal forms only… | ||
| The node_modules Iceberg: Mysteries Below the Waterline | Chris DeMars | 25m |
Claims that 60.4% of incidents happen when patches exist but aren’t applied and that transitive dependencies hide most supply chain risk, illustrated by the… | ||
| Inside MCP: Live Protocol Messages, Real-Time Flows, and Smarter Agents | David Parry | 23m |
Builds a working MCP server in Java using plain JSON over stateless IO protocol, demonstrates an MCP client inspector that discovers tools, runs synchronous… | ||
| What Happens When the Music Starts? The Architecture of Browser Audio | Kerline Moncy | 22m |
Three levels of browser audio are compared with concrete tradeoffs: the HTML5 audio tag gives no signal access; the Web Audio API exposes AudioNodes so you can… | ||
| Building Rapid Web Applications with Laravel + Filament | Aaron Krauss | 25m |
Combining Laravel (v13) with Filament (v5) claims to cut boilerplate: the demo app Simple Slides is built with Eloquent queries, Inertia front end, background… | ||
| Counting exercise repetition using live camera video | Alexey Inkin | 23m |
Uses MediaPipe pose landmarks to read 33 body points, reduces them to 16 metrics (8 bones × X/Y) and runs PCA to find the dominant motion axis, then detects… | ||
| AI Memory Is Broken And How I Fixed It | Chris Sean Dabatos | 24m |
Built a voice AI agent after MeZero dropped names and GBT4‑mini hallucinated family details, then shipped a three-layer memory that actually works: a… | ||
| Skip the Middleman: Direct, Secure UI-to-Agent Communication with Amazon Bedrock AgentCore | Andres Moreno | 14m |
Shows a working demo of a browser connecting over a pre-signed websocket URL to Amazon Bedrock AgentCore runtime, streaming responses token-by-token to the UI… | ||
| Beyond SEO: Optimizing Your Web Content for the Age of AI | Nerando Johnson | 25m |
Argues that traditional SEO no longer guarantees AI visibility and defines Generative Engine Optimization (GEO) with five differences from SEO, then gives a… | ||
| Open Source Voice AI: From ChatGPT's Infrastructure to Your App | Jesse Hall | 24m |
LiveKit shows how ChatGPT's Advanced Voice Mode runs on an open source real‑time stack, with a working hotel‑booking demo that fires async tool calls so the… | ||
| The Feature Nobody Used & What It Taught Me About Product Engineering | Bethany Ann | 24m |
A team spends a year building a feature that gets shut down because nobody uses it, and market data backs it up: Pendo reports 80% of features sit unused and… | ||
| From Internal Innovation to Hilton.com: Building Hilton’s First Agentic AI Travel Planner | Paul Coroneos, Raghu Mathukumalli | 19m |
Internal 1.0 ships January 2025, stalls after user tests, then a 48-hour hackathon in summer 2025 reimagines the product as a chat married to search pages; the… | ||
| Advanced RAG Systems: Scaling Retrieval-Augmented Generation in Production | Pranav Puranik | 25m |
Covers concrete recipes: use BM25+vector hybrid retrieval to return ~20–50 chunks, then run a cross‑encoder reranker on ~20 candidates to reduce to top 5–10… | ||
| 20 mins to Production: A Live AI-Build Challenge | Jerry Reghunadh | 26m |
Live builds a habit-tracker app using Claude and Gemini agents, a precreated NextJS on Vercel repo, and browser SQLite for local storage; agents generate a… | ||
| Are MCP Servers Dead? | Chris Griffing | 25m |
Argues MCP servers aren’t dead and gives concrete fixes and features: a stateless V2 spec released recently adds multi‑roundtrip calls, cachable tool lists… | ||
| The Three Ways AI Agents Fail. Here's How to Fix Yours. | Moeez Khan | 20m |
Enterprise GenAI projects hit silent failures at three concrete layers: tool contracts that return wrong or empty data (MIT NANDA 2025 finds 95% deliver no… | ||
| Why Your Bug Rate Isn’t a Code Problem: Engineering for Unstable Work Intake | Eric Bible | 22m |
Claims that defect rate follows intake stability, supported by anonymized JIRA data: Q1 hits 4 stories to 1 bug, Q2 degrades to 2:1 when intake is unstable… | ||
| Tools in your AI's Toolbox : An introduction to MCP Servers | Tim Rayburn | 25m |
Introduces Anthropic’s Model Context Protocol as a stateless HTTP client‑server that exposes methods described by JSON schema so models discover and invoke… | ||
| WTF, Rust for Web Development?? | Francesco Ciulla | 25m |
Rust proves practical for backend and full‑stack work with concrete wins: Cloudflare’s Pingora reimplements an HTTP proxy in Rust and serves one trillion… | ||
| Give Your Agent a Second Brain (and a UI to Match) | Bree Hall | 22m |
Builds a typed second brain in Sanity, queries it with Grock and uses Claude Haiku to map records to React components so agents return UI not blobs. Demo shows… | ||
| Build your First MCP App | Nick Taylor | 26m |
Remote MCP servers can return interactive UI not just text using the MCP Apps extension; the demo live-codes an app from a TypeScript React/Tailwind template… | ||
| The Claw: Minions of Toil | Joel Hooks | 23m |
Builds an always‑on agent cluster running on a Mac Mini or DigitalOcean droplet using OpenClaw principles and a moldable Pi harness, wires Fable instances for… | ||
| Herding Engineers: What I learned from Horse Trainers In Northern Texas | Alexander Crettenand | 23m |
Five concrete rules drawn from horse training map to engineering leadership: treat engineers like 1300‑pound animals you influence not control; train reactions… | ||
| Ship It First, Fix It Later: How a Medical Crisis Led to a Global Tool | Roxy Rodriguez Becker | 23m |
A niche G‑tube rate calculator starts as 71 lines of JavaScript and becomes a production React + TypeScript PWA with Tailwind, offline support, Supabase auth… | ||
| Documentation in the Age of AI: Why Docs Are a Product | Chris Nowicki | 23m |
Migrates Commerce documentation for BigCommerce, Makeswift, and Feedonomics from a legacy platform to Fern on a hard launch schedule and rebuilds docs as a… | ||
| Software Developer To CTO: AMA! (Ask Me Anything!) | Caleb Jenkins | 28m |
A checklist of non-code risks and tradeoffs: company runs ~200 people with ~11 in IT and treats every architecture choice as a purchase order, so a vendor… | ||
| Nobody is hiring vibe coders. | Dennis Ivy | 22m |
Timeline runs from 2023 ChatGPT through 2024 layoffs to the Opus 4.5 winter surge and a claim that AI now writes most code at some frontier shops. About 14–16… | ||
| Stop Prompting, Start Looping: Principles of Loop Engineering | Derrick Bozkurt | 22m |
Describes a loop architecture that runs an agent headless under deterministic guards, commits each tick, uses cheap iteration gates, a reviewer LLM, and hidden… | ||
| Deploy Your Own VideoGen Model | Patrick Hulce | 26m |
Walks through training a DiT/LTX-2.5 pipeline using LoRA fine-tuning on just 20 five-second clips, runs on Modal with B200 per-second billing, and deploys a… | ||
| Make Your Serverless Functions Do More with Durable Executions | Erik Hanchett | 24m |
Shows an AWS Lambda durable executions pattern with a working e‑commerce demo that authorizes a Stripe payment as uncaptured, pauses for manual admin approval… | ||
| Gotta Feed the Beast - Data Engineering at Scale | Claire Bourdon | 24m |
Data engineering means building and operating the pipes that feed humans and AI agents with usable data at scale not theory but concrete tradeoffs and costs… | ||
| Boost Developer Productivity with Buildpacks | Joey Brown | 25m |
Cloud Native Buildpacks replaces ad hoc Dockerfile app builds with a spec-driven pipeline that uses pack (or the CNB lifecycle) to detect language, run modular… | ||
| Chaos to Control: Automating the Identity Life Cycle in the Cloud | Noah Jenkins | 24m |
Cloud identity lifecycle automation that actually works: use Workday as the source of truth, trigger Azure Functions or Logic Apps to provision Entra ID… | ||
| Never Click Apply. Get Hired. | Leon Noel | 25m |
A repeatable hiring system claims 11x better conversion than cold apply and says cold easy-applies yield about two offers per 1,000 submissions; it replaces… | ||
| Stop Writing the Analysis Script: Let the Data Decide What to Look At | Stacy Haven | 27m |
A manual trading indicator built from ~75–79 million rows and 1,180 trading days becomes a fast experiment when fed to Claude rather than hand‑coding rules… | ||
| Serverless API Gateways for High Scale Systems Performance Cost and Security | Sujeeth Reddy Pasham | 24m |
Serverless API gateways scale from idle to thousands of concurrent executions and cut infrastructure waste by tuning memory and CPU (examples: fixing memory… | ||
| When AI Gets Manipulated: Security Risks in LLM Applications | Vishnu Ramineni | 26m |
Generative AI systems accept both system prompts and user input and remain vulnerable to prompt injection that overrides guardrails; the demo uses an… | ||
| Unlocking CI Performance in the Age of Agentic Development | Scott McAllister | 24m |
Agentic commits multiply PRs and expose CI waste: GitHub reports 46% AI-generated code and 1.2M Copilot PRs/month, pipeline overhead often hits 58% of run… | ||
| Gating Cognition: How Modern AI Architectures Minimize Runtime Reasoning | Hunter Long | 26m |
Compares observed ReAct agent flows that incur four or more LLM calls and tool descriptions in every prompt with a ReWOO planner that outputs a JSON plan… | ||
| Secure Coding Meets Critical Infrastructure Defense in Water Systems | Venkata Kartheek Reddy Somasani | 25m |
Municipal water systems at City of Dallas scale move from air‑gapped PLCs to cloud dashboards and remote vendor access, increasing entry points; real incidents… | ||
| Design Systems 101: The Developer Approach to Better Frontends | Kaleb Garner | 21m |
Builds a JavaScript design system around React, Storybook, CSS tokens, starting from primitives (box, stack, typography) up to layouts and composable… | ||
| Is Beauty Only Skin Deep? | Noah Harris | 19m |
Claims that clean architecture buys back engineers' attention and shows it with a concrete JavaScript example: a Stripe order webhook that must do three things… | ||
| rm -rf: Horror Stories From Unsandboxed AI Agents (and How Docker Fixes This) | Rishab Kumar | 26m |
Between July 2025 and February 2026 documented incidents include a Claude Code session that ran terraform destroy wiping 2.5 years of student data for 79,000… | ||
| Zero to Production: A Simple Technical Framework for Low-Cost Scalable Infrastructure | Whittington W | 25m |
A practical four‑question framework surfaces exact risk: name the meter, dial, ceiling, feedback for each subcomponent and calculate how many billable units it… | ||
| Your Prompt Is Not a Unit Test — instrumenting LLM apps in production | Kyle Taylor | 29m |
Shows how to instrument LLM features end to end: capture OpenTelemetry spans and trace UI → agent → Redis/Postgres → model calls (Azure OpenAI, Bedrock) to… | ||
| Give Your SDK Superpowers: Building AI Tools for Your Developers with Skills and MCP | Brian Munz | 24m |
Shows that combining Claude Skills and the Model Context Protocol gives AI structured SDK knowledge plus real tooling access so generated code actually… | ||
| Weights vs. Words in Fine-Tuning & Enhancing Agentic AI Workflows | Reece Iriye | 25m |
Compares LoRA fine-tuning against smarter prompts and software scaffolding by building a restaurant search engine using a react loop harness with three tools… | ||
| In the age of AI: if it's not Accessible, it's not Acceptable | Amanda Mayfield | 22m |
Domino's accessibility case reaches the Supreme Court denial in 2019, web lawsuits climb from 2,000 (2018) → 5,000 (2025), and settlements run $60k–$285k plus… | ||
| When Vectors Aren’t Enough: Unlocking AI’s Hidden Connections with Graphs | Nyah Macklin | 27m |
Argues that vector-only RAG misses multi-hop, relational queries and shows a hybrid architecture: a knowledge graph (Neo4j) paired with a vector store and… | ||
| Data Governance in the AI age | Miguel Guillén | 24m |
Claims that untraining an LLM is practically impossible except by full deletion, that poisoning or “poison pill” retraining causes model degradation, and that… | ||
| Revolutionize Your Backend: Mastering Server Actions with Next.js and React | Avindra Fernando | 25m |
Next.js server actions convert simple JS functions into backend endpoints, demonstrated by an addUser function that writes to a Postgres database and initially… | ||
| Property Based Testing for the Browser with Bombadil | Nate Custer | 26m |
Open-source Bombadil runs property-based tests written in TypeScript/JavaScript against a real browser tests session, generating weighted random actions… | ||
| Zero-Trust Architecture for Agentic AI: Securing Multi-User Access and Third-Party Integrations | Borko Djurkovic | 23m |
An enterprise agent platform enforces OIDC SSO for identity, decouples authorization from app code using OpenFGA relation-based ACLs, and evaluates permissions… | ||
| Your Browser Has Trust Issues: Why Local Devices Break Modern Web Apps | Chris Edgington | 25m |
Modern browsers block cloud HTTPS pages from talking to local hardware unless four security gates are satisfied: mixed content (HTTPS→HTTP blocked), CORS… | ||
| The Playwright Cipher: Writing Tests Like a Rap Verse | Wayne Jones | 25m |
A live Playwright demo builds an album‑art review site test suite and shows four concrete scenarios: a smoke test that verifies homepage and headings load, a… | ||
| From Guesswork to Governance: Bringing API Discipline to Data Streaming | Sandon Jacobs | 25m |
Claim: move validation upstream so producers can’t write poison data and consumers can trust event formats. Uses schema registry (stored in Kafka)… | ||
| Trust, But Verify: Human-in-the-Loop for Agents That Actually Matter | Michael Liendo | 23m |
Maps human-in-the-loop patterns to concrete mechanisms: lightweight inline confirmations for seat or preference questions, token vault and OBO for long‑lived… | ||
| Comparing Promises, Observables, and Signals | Lance Finney | 26m |
Compares concrete behaviors: Promises fit single-value HTTP use cases and show a fetch example with fetch + AbortController for cancellation; Observables… | ||
| From Prompt to Prod: Build and Deploy with ZCP (Work Shop) | Francesco Ciulla | 01h 13m |
Workshop builds and deploys a working multi-service app on zerops hosting using an AI coding agent and a pre-baked recipe. Deployment runs agents tied to BYO… | ||
| Teach Your To-Do List to Listen (Workshop) | Corey Weathers | 01h 25m |
Hands-on workshop walks a JavaScript/Node to-do app through voice: it wires the Deepgram Voice Agent API for transcription, an LLM for intent, and Deepgram… | ||
| Close the Tab: Build Jobs That Keep Running (Work Shop) | Ojus Save | 01h 26m |
Deploys a demo stock research agent that first ties work to the HTTP request so closing the tab loses progress, then replaces that with Render Workflows so… | ||
| nothing matches | ||