Scaling Vision-Heavy Kimi-VL with Heterogeneous E/PD on llm-d and SGLang #llmd #ai
llm-d.ai
Scaling Vision-Heavy Kimi-VL with Heterogeneous E/PD on llm-d and SGLang | llm-d
twp.ai
The guy at the end of Bruno’s leash
@davidgs.com
DevRel leader since forever, now Community nerd. Huge 💜, 🌶️🧠, dog lover, Rottweiler Rescuer, clumsy coder. Always kind, not always nice. Was @davidgsIoT on the other place. Rottfather. Friend of Bill’s. 🏡davidgs.com
Securing Branch Networks With Firewalls, VPNs, IDS/IPS, and Identity-Based Access
feeds.dzone.com
Securing Branch Networks With Firewalls, VPNs, IDS/IPS
twp.ai
Oh darn, this chud has exited the election. What a creep. apnews.com/article/chuc...
North Carolina congressman Edwards quits reelection bid after committee recommends his censure
North Carolina Republican congressman Chuck Edwards has withdrawn from his reelection campaign days after the House Ethics Committee recommended he be censured for unprofessional and inappropriate con...
apnews.com
Performance Testing With JMeter Beyond the Basics: Distributed Load, Realistic Profiles, and Identifying Security Bottlenecks
feeds.dzone.com
Performance Testing With JMeter Beyond the Basics
twp.ai
Happy to announce the release of Duckling UI v0.0.14 today! Biggest feature is DocLang export format: Export conversions to .dclg.xml DocLang is the future of machine-optimized document formats. Docs: https://twp.ai/E5EwGq Source: https://twp.ai/4huRsi
KubeCon + CloudNativeCon India 2026: Our Kubeflow Community Experience
KubeCon + CloudNativeCon India 2026: Our Kubeflow Community Experience | Kubeflow
Introduction
twp.ai
Serving GLM-5.2 for Agentic Workloads on llm-d #llmd #ai
llm-d.ai
Serving GLM-5.2 for Agentic Workloads on llm-d | llm-d
twp.ai
Intelligent Inference Scheduling with llm-d #llmd #ai
llm-d.ai
Intelligent Inference Scheduling with llm-d | llm-d
twp.ai
llm-d 0.4: Achieve SOTA Performance Across Accelerators #llmd #ai
llm-d.ai
llm-d 0.4: Achieve SOTA Performance Across Accelerators | llm-d
twp.ai
Hey @github.com and @datadoghq.com , did you know that some of your employees are going to speak at this conference in September? Maybe rethink that.
Carmen Huidobro@hola-soy-milk.online · last wk.Just emailed the orgs, I can't MC a conference that condones this behaviour. I'm devastated. 💔
Why Enterprise AI Agents Fail: A Runtime Data Governance Pattern for Reliable Answers
feeds.dzone.com
Why Enterprise AI Agents Fail
twp.ai
KV-Cache Wins You Can See: From Prefix Caching in vLLM to Distributed Scheduling with llm-d #llmd #ai
llm-d.ai
KV-Cache Wins You Can See: From Prefix Caching in vLLM to Distributed Scheduling with llm-d | llm-d
twp.ai
llm-d 0.2: Our first well-lit paths (mind the tree roots!) #llmd #ai
llm-d.ai
llm-d 0.2: Our first well-lit paths (mind the tree roots!) | llm-d
twp.ai
llm-d 0.3: Wider Well-Lit Paths for Scalable Inference #llmd #ai
llm-d.ai
llm-d 0.3: Wider Well-Lit Paths for Scalable Inference | llm-d
twp.ai
llm-d Week 1 Project News Round-Up #llmd #ai
llm-d.ai
llm-d Week 1 Project News Round-Up | llm-d
twp.ai
llm-d Community Update - June 2025 #llmd #ai
llm-d.ai
llm-d Community Update - June 2025 | llm-d
twp.ai
llm-d v0.7: From Feature Introduction to Production Hardening #llmd #ai
llm-d.ai
llm-d v0.7: From Feature Introduction to Production Hardening | llm-d
twp.ai
llm-d 0.5: Sustaining Performance at Scale #llmd #ai
llm-d.ai
llm-d 0.5: Sustaining Performance at Scale | llm-d
twp.ai
No Kubernetes? No Problem: llm-d Now Runs Anywhere #llmd #ai
llm-d.ai
No Kubernetes? No Problem: llm-d Now Runs Anywhere | llm-d
twp.ai
This is exactly the kind of bullshit spin that caused me to turn off NPR (Nice Polite Republicans). “May violate”? It’s blatant corruption. Call it that.
NPR@npr.org · 5d agoThe president's media company is offering investors early access to his Truth Social feed for up to $100,000 a month. Experts say the service may violate insider trading laws.
Free idea: start a replacement federation to FIFA but call it FAFO. I’ll let y’all figure out what the acronym stands for.
Turned on @npr.org this am for the first time since the election only to hear them describe Trump’s Truth Social API as another “interesting way” for the Trumps to make money. It’s not interesting. It is naked, blatant corruption. Not calling it that is disgraceful.
Production-Grade LLM Inference at Scale with KServe, llm-d, and vLLM #llmd #ai
llm-d.ai
Production-Grade LLM Inference at Scale with KServe, llm-d, and vLLM | llm-d
twp.ai
Heterogeneous inference serving across three GPU vendors with llm-d #llmd #ai
llm-d.ai
Heterogeneous inference serving across three GPU vendors with llm-d | llm-d
twp.ai