[javaone 2026] Developing an Asynchronous Application with Virtual Threads and Structured Concurrency

Posted on by
Reply

Speakers: Jose Paumarrd & Ana-Maria Mihalceanu

See the live blog table of contents

Structured concurrency pattern. Note tasks end by autoclosable block complete

try (var scope - StructuredTaskScope.open()) {

  var subtask = scope.fork(lambda)

scope.join()

var result = subtask.get();

}

Scopes

  • A scope can spawn other scope
  • Closing the parent scope closes the child scopes

Lab

https://github.com/JosePaumard/2026_JavaOne-Loom-lab/blob/Step-00_Initial-application/JavaOne-Loom-Lab.md

[javaone2026] Running GPU-Accelerated AI Inference from Java at Uber Scale

Posted on by
Reply

Speakers: Baojun Liu & Anshuman Mishra

See the live blog table of contents

Michelangelo

  • Uber’s unified ML platform
  • 20K models trained/month
  • 5.3 models in prod

Java

  • Online prediction is all Java
  • High concurrency orchestration
  • Business logic management
  • Production ecosystem integration

Spark ML pipeline

  • C++ is ML runtime and Math libraries
  • Java manages threads, coordinates execution, handles back pressure

Scaling

  • Model sizes have increased from 100K to 800G
  • Traffic growth about 30% per year

GPU/CPU

  • GPU cluster – 20x cost reduction and order of magnitude less instances
  • NVIDIA Triton inference server (C++)
  • GPU can handle 10x-100x more requests
  • Dynamic latency – trades latency for throughput
  • Scale down instance count for large models to save memory and up for high queries per second
  • One GPU to multiple CPUs. CPU bottleneck – cold start, error spikes, memory pressure

Techniques

  • Profiling
  • Send dummy data to avoid cold start
  • Tune GC memory and pause time
  • Frequent efficient collections
  • Panama
  • Virtual threads
  • Upgrade Java newer version to reduce CPU use

Other notes

  • Java is not the bottleneck, bad architecture is

My take

The beginning was the same slide as the keynote. Ok to review though. It got new quickly and it was nice to see what Uber is doing. Good questions from the audience

[javaone 2026] fundamentals of software engineering in the age of ai

Posted on by
Reply

Speaker: Dan Vega (@therealdanvega)

Nate Schutta (@nts.bsky.social) – not here due to snow

See the live blog table of contents

General

  • They wrote a book called Fundamentals of Software Engineering

History

  • 1950s: Punchards/machine code
  • 1960s: Assembly language
  • 1970s: High level languages (ex: COBOL)
  • 1990s: OOP and frameworks
  • 2010s: Cloud/platform
  • 2020s: AI assisted development

History repeats

  • Each time, here that engineers will be replaced by IDEs, co-code, etc. Still here
  • Roles change

Role of dev

  • We don’t sit and write code all day.
  • Writing code is one of the tasks
  • ex: problem solving, communication, debugging
  • SDLC – Plan, Analysis, Design, Implementation, Testing, Deployment, Maintenance
  • Outside world thinks it is just about code
  • Even vibe coders plan

Hiring/News

  • Market overhired
  • Always ups and downs in job market
  • Many large layoffs in 2025
  • “”Job hugging” vs “job hiring” – https://www.cnbc.com/2025/08/18/job-hugging-job-hopping.html
  • Salesforce regrets firing senior engineers and replacing with AI
  • Amazon had prod deleted by AI without sufficient oversight
  • Anthropic said a year ago that AI would write 90% of the code in 3-6 months. Keep in mind source of claim. Also they have access to “free tokens” every day. Also they are hiring a number of engineers.
  • 2015 – self driving cars in 2 years, 2016 – radiologists replaced in x years. Now radiologist driving own car to work.

Overhyped claims

  • Tool makers – want to sell tools
  • Non programmers – lowered barrier to entry is good; but don’t know all it entails.
  • Business – Blaming AI is a good excuse.

Vibe coding

  • Great for weekend programs. At work, more meticulous about what goes into Prod
  • “We vibe code a 30K a month SaaS app in 64 minutes” – click bait. Can clone screens but not getting 30K
  • Worlds apart from engineering system

AI

  • “Lowering the Floor and Raising the Ceiling: Building With AI”
  • Give us superpowers
  • Not a silver bullet

Social

  • Made website and then link to c drive
  • AI broke code/repo
  • AI generated code and don’t understand it

Other problems

  • Who is supporting the code that AI broke
  • In danger of stopping pipeline that creates experts

Exciting time to be a builder

  • Find joy in solving problems and making something that works
  • Something you imagined became real
  • Not about typing
  • Less abandoned ideas for side projects. Easier to do just for fun

Experience

  • Anyone can create code; can they create software.
  • Wouldn’t higher neighbors kid to as photographer for a wedding
  • In sports, constantly work on fundamentals.

Paths to computer engineering

  • college degree – undergrad prepare you for graduate program – os, compiler theory, etc
  • bootcamp – lots of info in very short time – frameworks, language de jour, debugging, mostly learn how to code
  • self taught
  • But there is a huge gulf between any of these and what you need to know to be successful.
  • Approaches even out over time
  • Coding is more about communicating than computing. Language aptitude better predictor of learning to code; not math.
  • Problem solving and curiosity key

Reading code

  • Spend 10x more time reading code than writing it
  • With human languages, we learn to read first.
  • In era of AI, spend even more time reading code
  • AI can write code quickly but needs constant supervision
  • “What idiot wrote this code?” – oh, me from months ago
  • Shipping speed affects readability
  • IKEA effect – place a higher value on things we created

Learning more languages

  • Easier to learn other languages/frameworks when know more
  • More receptive of new features

Managing your career

  • “The only way to do great work is to love what you do. If you havne’t found it yet, keep looking, Don’t settle” – Steve Jobs
  • When start as a software engineer, think role is to learn to code, get better at coding and keep coding. But tech offers other unique roles, not just manager
  • Advocate for yourself; your manager can’t read your mind
  • Make new friends/networking.
  • There’s always something you know more about than something else, share
  • Learn from others
  • Find what you are passionate about – ex: manager, dev advocate, sales engineer

Deliberate skill acquisition

  • Work backwards from where want to be
  • Need to accept than can’t learn everything
  • Must learn core skills – ex: data structures, design patterns
  • T shaped – horizontal is general knowledge and vertical is deep knowledge in primary stack
  • Build a personal technology radar
  • Record your wins – provides confidence and helps track professional growth. More useful if specific (metrics, tech skills)

AI

  • Ok to feel overwhelmed
  • Too many new things “supposed to learn”

AI Dev Stack

  • Models – a new one comes out every week. Experiment and see what works
  • Context and memory
  • Tools and functions
  • Agents and workflows
  • Your app

Will AI replace software developers

  • It depends
  • Software engineering isn’t what it was two years ago
  • Need to learn things and adapt to current events

Takeaways

  • Keep your passion
  • Technology changes constantly
  • Don’t put head in the sand
  • Define self by problems will solve in the future vs what did in past
  • Whatever path you choose, change is inevitable
  • You are responsible for your own career
  • Fundamentals will always serve your well
  • “Excellence is doing ordinary things extraordinary well”

My take

I was torn on whether to go to this session because I wasn’t clear on whether it was current events or covering the fundamentals like design that I’ve been doing for 20+ years. I was pleased it was the former. It was fun. Some events I noticed go by and some I didn’t. Fun start to the morning. Also good tie to the actual fundamentals rather than teaching them.