ThrillhouseBot v0.7.1 just arrived! 🔥
Less noise in AI code reviews:
- Fixed findings now resolve their threads
- Better duplicate detection
- Fewer security scan false positives
- PR summaries grounded in actual code
Open source & self-hosted.
https://t.co/tG2mUR8R1S
#OpenSource #CodeReview #Java #Quarkus #Langchain4j
Out now: A new version of quarkus-opentelemetry-exporter! 🚀
v3.40.1.0 can send OpenTelemetry data directly to Azure, GCP, and Sentry with its dedicated exporter support.
This version is based on @QuarkusIO 3.40.x LTS.
https://t.co/fO1m06cq27
#LangChain4j 1.21.0 is out! 🚀
Meet Decision Models: instant yes/no, choice and scale answers with probabilities.
Decision Services turn any Java interface into a classifier.
Plus model routing, re-ranking, tool selection, guardrails and agentic routing!
https://t.co/LP02Y1ogQu
ThrillhouseBot 0.7.0 is out 🚀
Secret/IaC scanning, CI + linked issue context, review learnings, smarter per-call prompt routing, webhooks, and much cleaner follow-up reviews.
All opt-in. No new GitHub App permissions.
https://t.co/AFgYGQtoFT
#CodeReview#DevTools#Quarkus #Java #Langchain4j
Sometimes you don’t need another presentation. You need a conversation.
That’s the #JCON Mentorship Roundtables at the #IBMTechXchange 2026.
Find a topic. Walk up. Ask questions. No signup.
Visit us at the Georgia World Congress Center in Atlanta!
👉 https://t.co/ehnrEwFUlx
#Java #JCONUSA26
Quarkus 4 beta 1 is out!
Take a look at the new @micrometerio Observation API support and the Observability dashboard in the Dev Ui.
OpenTelemetry defaults changed:
- Exporter now uses HTTP,
- Sampling is now set to 10%
… let us know what you think.
After two years of effort, we’re thrilled to announce Quarkus 4.0.0.Beta1 is now available! Give it a spin, migrate your extensions, and share your feedback, it’s much appreciated. We’re targeting the final release by the end of November. https://t.co/jpr2zM2scy
My small contribution to these decision model world… https://t.co/EvRcWGHGs1 it’s based on @googlegemma 4 e2b (5b MoE 2b active), fine tuned using RLCD that delivers structured decisions in a single forward pass, trained locally using 2xRTX 4060ti 16GB
#gemma#machinelearning #pytorch #llm #openweight
One Java. Two rooms. Three days.
From Oct 27–29, the #JCON Rooms bring developer-to-developer #Java sessions to the #IBMTechXchange 2026 conference in Atlanta.
Real problems. Useful ideas. A lot of Java & #AI
More information at: https://t.co/yK7mS2rWdP
#JCONUSA26
v0.6.7 is out and it's smarter than ever. Using DeepSeek-V4-Flash, it catches almost every common coding problem in the most-used languages available today, costing 2 cents per review.
https://t.co/FcYPGYNOkc
#java#quarkus#langchain4j#github#ai
Meu novo artigo acaba de ser publicado no @foojayio! 🚀
“From Spec-Driven Development to Living Specifications in Java Projects”
https://t.co/2iSIxcL5DD
Ah, sim, amanhã também levo essa discussão ao #DevConverge SP 2006 com o SDD4J. Nos vemos lá!
#Java#SDD4J#AI