Exciting day!!! The Early Release of our new O’Reilly book is out! First two chapters. More to come soon! Grab it while it’s hot! learning.oreilly.com/library… #Java #Agents @OReillyMedia @alexsotob @mariofusco

MCP servers: because what your attack surface really needed was a plugin ecosystem written over a long weekend by strangers. Zero audits, infinite trust.

Why searching for “Nesbo” returns nothing when Jo Nesbø is right there in the index. A classic Hibernate Search analyzer mismatch, and the small asciifolding change that fixes it. Hands-on diagnosis + the regression test that keeps it fixed. the-main-thread.com/p/hibern… #Java #HibernateSearch #Quarkus

I added a transactional outbox to a Quarkus checkout: order and event commit together in PostgreSQL, then a scheduler publishes to Kafka. The walkthrough covers JVM crashes after commit and after Kafka acknowledgement, plus an idempotent fulfillment insert. I also explain where Debezium's embedded Quarkus engine fits. the-main-thread.com/p/quarku…

👇🏻 this!

I connected #LangChain4j to Apple's local fm serve endpoint with a few #JBang scripts. On my M4 Max, five repeat requests had a median of 41.52 output tokens/sec, including the initial wait. Two simple instruction checks passed; a conflicting instruction and the tool-call check failed. I walk through the #Java code and what happened: the-main-thread.com/p/apple-…

I’m pretty sure that there’s more iPhone Duo cases out there than actual phones 🤣

Prod credentials plus an agent equals a very expensive way to find out that you never tested your backups.

Code Review After Code Gets Cheap oreilly.com/radar/code-revie… @radar

How small can a Java container image realistically be? Containers with Java apps have the reputation of being bloated. How true is it today? Can a typical framework-based Java REST service fit into a 50MB container? And 30MB? Watch my experiment: piped.video/2f0LUg5_GvQ

The AI hype cycle: a thousand startups wrapping the same API, praying OpenAI doesn't ship their feature on Tuesday. Schadenfreude is the only thing in this industry that scales.

If you have high standards, you will inevitably offend people at some point in your work or life. You are responsible for your delivery, but not how they react or respond. Hold the standard anyway.

For this Main Thread Podcast episode, I retrace MOSS-TTS, VoxCPM2 and F5-TTS, then Qwen3-TTS with a local LoRA adapter on my MacBook. Reference audio and chunk boundaries changed the pace. I compared the clone, a speaker-only control and the adapter under the same settings. Listen: the-main-thread.com/p/findin…

"Agentic coding will 10x developers." It does. It 10x's the code, the bugs, and the number of people who have to pretend they understand the diff.

JEP 544 (openjdk.org/jeps/544), the remaining part of Leyden has landed in JDK mainline. So now we can do this! This source launcher involves javac compilation, that is the bulk of the work here.

All the #Devoxx Belgium 2026 talks are now available on our YouTube channel. Enjoy and share the knowledge! 🤩 piped.video/channel/UCCBVCTu…

I built a small Quarkus 3.39.4 app to check MDC across async calls. With ThreadContext.withContextCapture(), the common-pool continuation still read null. Returning it to the saved Vert.x request context restored the ID; the external stage's own MDC stayed empty. The tutorial includes ManagedExecutor and Mutiny controls, failure checks, and concurrent HTTP tests against the packaged JVM app. the-main-thread.com/p/quarku…

My session with @mariofusco at @Devoxx "A Year of Agentic AI Evolution: Lessons Learned Building Production grade Agentic Systems" is available to stream: piped.video/nEW0kosJjc8?si=GSC4… @langchain4j @QuarkusIO @IBM #IBMTechVoices

Dusting off this tweet from April because this gap in shared understanding of LLM capability is *widening*. It's now less "two groups of people speaking past each other" and more a sharp funnel. - Napkin math somewhere around 6B people (~75% of the population) have barely come in contact with LLMs at all. - Around 1-2B (~20%) are casual and infrequent users of free-tier ChatGPT-like products. This group sees derpy chatbots and treats them a bit like a better Google search, a writing aid, or etc. Maybe an agent tries to book you a flight. I have non-tech friends who (reasonably, imo) say they have not much use for it at all. Even many professionals outside of math&code are in this tier. For example, execs and many other functions spend a lot of their time talking to other people, so while they understand what is happening intellectually, it is still second-hand and a bit abstract. - Now we get to professional use of frontier-grade LLMs in math&code. Somewhere around 20M people (0.2%) see first-hand that large, complex projects that used to take them weeks/months can now be completed by agents with a prompt. Building apps, copying apps, translating apps, decompiling apps from binaries... This has all happened very quickly and recently - less than 1 year ago, I was writing code manually by hand, typing memorized computer code commands into a code editor character by character, occasionally pressing Tab to autocomplete a little chunk of code. - And finally we get to the ~5,000 people (~0.00006%) with access to frontier-grade systems internally. The external world has seen the preview. It looks like swarms of thousands of agents collaborating over weeks on software mega projects: minting zero days, running cyber attacks and defenses at machine speeds, discovering new science, advancing the frontier of mathematics. Things that would have taken top professionals in the industry years of work. Meanwhile, human review and comprehension are starting to fall behind. For example, people are still involved in the "archeology" of the OpenAI-HF incident from many months ago. Mathematicians may be poring over the 722 manuscripts on frontier mathematics for a while. The funnel is driven by a combination of factors: 1. The impact scales with ambition, problem size, and horizon. A question with a paragraph answer barely stresses the system. You need a reservoir of big, difficult problems that you really care about. This aspect drives the consumer / professional dimension of the funnel. 2. The jaggedness of the system (which I have written about a lot separately). Capability peaks in domains that are digital, verifiable and economically valuable. This is because LLM capability emerges from reinforcement learning on verifiable rewards on a curated environment mixture driven by revenue potential. This aspect primarily drives the area (e.g. math&code) dimension of the funnel. 3. Access. Free-tier, paid-tier, internal. So this is the weirdness of the moment. The general public has mostly not interacted with these systems. When they have, it looks like a derpy chatbot. The majority of professionals still see only a modest uplift. And a small sliver of professionals are experiencing the vertigo of the curve going vertical. And it is all happening at the same time.