Spring Office Hours

Join Dan Vega and DaShaun Carter for the latest from the Spring ecosystem. We’re back from Dev2Next with a recap. Dan shows off two things he built last week. One is Kiln, a warm new theme for IntelliJ. The other is a set of demos for the JSON API coming to JDK 28. Then we dig into a busy week of model releases. Meta released the weights for Muse Spark Open, NVIDIA fit a 74B model into 24GB, and DeepSeek shipped a faster V4.1 Flash. Local inference is getting real, but RAM and VRAM prices keep climbing. We’ll talk about when running models on your own hardware makes sense, and how to point Spring AI at a model on your own machine. We’ll also answer your questions live.
Show Notes:
dev2next Conference
Spring AI: There and Back Again (demo code)
ArchUnit
JEP 540: Simple JSON API (Incubator)
Native JSON demos for JDK 28
Kiln: a warm theme for JetBrains IDEs
Spring AI Evaluation Testing
Spring AI Recipes by Craig Walls
The show:
springofficehours.io - episodes, schedule, community
Spring Developer on YouTube - join us live every Monday
Dan Vega
DaShaun Carter
The Spring Blog - the news we cover each week

Join Dan Vega and DaShaun Carter for the latest updates from the Spring Ecosystem. In this episode, Dan and DaShaun are joined by the legendary Josh Long. In this episode, we will look into the future of software development. We will also address at least one hot take that recently made the rounds. You can participate in our live stream to ask questions or catch the replay on your preferred podcast platform.
Show Notes:

Rails World 2026 Opening Keynote - DHH
JEP 540: Simple JSON API (Incubator)
jfmt: Netflix's formatter for Java
Spring Modulith
Qwen2.5-Coder 1.5B on Ollama
Ollama Cloud
Josh Long on YouTube (Coffee + Software)
joshlong.com
The show:

springofficehours.io - episodes, schedule, community
Spring Developer on YouTube - join us live every Monday
Dan Vega
DaShaun Carter
The Spring Blog - the news we cover each week

Jev is a new model from TypeSafe that answers typed questions about your content. You send it some text and the decisions you need. It returns a probability, a choice, or a score instead of a paragraph. That makes it a natural fit for the decisions your code already has to make. Is this ticket urgent? Which team should own it? Which model should answer this prompt?
In this episode, Dan and DaShaun talk about why this matters for Java and Spring developers. Most of us have been asking a chat model to make a decision and then parsing prose to get the answer out. Jev gives you a typed answer you can branch on, with a confidence you can check. Dan shares what he learned getting it running in Java and Spring Boot, and the two of them talk through where a decision model fits next to Spring AI and where it does not.
Then it is your turn. Bring your questions about Jev, Spring Boot 4, Spring AI, or anything else in the Spring ecosystem. We will answer them live.
Show Notes:

Introducing System One Models & Jev - TypeSafe AI's launch post and the article Dan started with: a decision model, not an LLM
Spring AI and TypeSafe Jev: Fast, Cheap, Structured Decisions - Christian Tzolov's post from the morning of the show, with the spring-ai-starter-typesafe dependency, JevJudge, guardrail advisors, and seven runnable examples
spring-ai-typesafe on GitHub - the Spring AI Community project behind that starter: a Java SDK for the Jev API plus the Spring AI integrations
Getting Started with Jev in Java and Spring Boot - Dan's write-up of the single-file Java example and the Spring MVC RestClient version
jev-spring-boot-starter - the Spring Boot 4 starter Dan built over the weekend: RestClient, properties for the API key and model, records for the request and response
How Spring Boot 4 Starters Actually Work (I Built One for Jev) - Dan's video on building that starter
It's Not an LLM, So What Does It Do? 3 Real Apps I built with Jev - the examples Dan described on the show, including the model router and the real-time writing checks
spring-ai-model-router - the Spring AI example where Jev classifies the prompt and picks which model should answer it
Spring release calendar - DaShaun's favorite place to check what ships on September 24 and the milestones on the way to Spring Boot 4.2 in November
I Publish My Podcast with Claude Code and Skills - Dan's video on replacing this show's editing and publishing workflow with skills
Fundamentals of Software Engineering - Dan and Nate Schutta's book, with the career development chapter Dan pointed to
S5E23 - Java 27 Release Party with Billy Korando - last week's episode
The show:
springofficehours.io - episodes, schedule, community
Spring Developer on YouTube - join us live every Monday
Dan Vega
DaShaun Carter
The Spring Blog - the news we cover each week

Join Dan Vega and DaShaun Carter for the latest updates from the Spring Ecosystem. In this episode, Dan and DaShaun are joined by Kansas City JUG organizer and Java Developer Advocate, Billy Korando. In this episode, we celebrate the release of Java 27. You can participate in our live stream to ask questions or catch the replay on your preferred podcast platform.
Show Notes:
JDK 27 - general availability September 15, 2026, with all nine JEPs
Java 27 Launch Stream - the Java channel's launch-day replay, with Billy's JFR demo and his interview with Roman Kennke
JEP 534: Compact Object Headers by Default - 10 to 20 percent less heap for most workloads, now on by default
JEP 523: Make G1 the Default Garbage Collector in All Environments - no more silent fallback to Serial GC in small containers
JEP 533: Structured Concurrency (Seventh Preview) - the last piece of Project Loom, aiming for final in JDK 28
JEP 531: Lazy Constants (Third Preview) - set once, later than final allows, with JIT optimizations to match
JEP 536: JFR In-Process Data Redaction - keeps API keys and passwords out of flight recordings before you hand them to a human or an agent
Transitioning Java to more frequent security updates - Oracle's plan to move Java security patches from quarterly to monthly, updated September 10, 2026
Project Babylon - code reflection and GPU programming from Java, DaShaun's pick for what to try next
S5E22 - Live from KCDC - last week's episode, recorded at the conference Billy helps organize
Billy Korando on LinkedIn - where Billy says to find him
Resources:
springofficehours.io - episodes, schedule, community
Spring Developer on YouTube - join us live every Monday
Dan Vega
DaShaun Carter
The Spring Blog - the news we cover each week

Join Dan Vega and DaShaun Carter for the latest updates from the Spring Ecosystem. In this episode, Dan and DaShaun broadcast live from the Kansas City Developer Conference, one of the largest community driven developer events in the country. Expect conference highlights, hallway track conversations with speakers and attendees, and the latest news from the Spring and Java worlds. You can participate in our live stream to ask questions or catch the replay on your preferred podcast platform.
Show Notes: 
Craig Walls, Spring AI Recipes index
habuma/spring-ai-recipes on GitHub
Spring AI Recipe: Running Against Local LLMs (Medium)
JDK 27 project page
KCDC
Spring AI Community on GitHub
Fundamentals of Software Engineering, danvega.dev