I'm hosting the second ever AAIF Toronto Meetup December 9 at the Cohere offices with @virtualized6ix @SebbyCorp and @DaleSeo !
Register here for a night of technical talks and networking abous OSS Agentic AI: luma.com/5lhxsfth
Register here for a night of technical talks and networking abous OSS Agentic AI: luma.com/5lhxsfth
1
1
1
108
@virtualized6ix @SebbyCorp @DaleSeo We are also looking for speakers! Have an interesting topic to share about OSS and agentic AI?
Submit a talk here: forms.gle/piJohUnMvcfN82…
Submit a talk here: forms.gle/piJohUnMvcfN82…
0
0
1
29
We just released github.com/containers/kub…, addressing a moderate security issue with the Kiali toolset.
If you use the Kiali toolset and do not have a hardcoded Kiali url in your config, please update or look at workaround steps!
github.com/containers/kub…
If you use the Kiali toolset and do not have a hardcoded Kiali url in your config, please update or look at workaround steps!
github.com/containers/kub…
2
0
1
28
Reports like this show why it is still valuable to understand the code, whether you use AI to create it or not
At HPTS, Kyle mentioned a company confidently handed him a vibe-coded database to test. Now Jepsen report on ParticleDB is out. IMO I feel sad for the engineers involved. It's a long report, but at least read the part on LLM tradeoffs.
"Perhaps engineers will begin to treat any new software project with suspicion—the signals we used to rely on as markers of a mature system are now easier to counterfeit."
jepsen.io/analyses/parti…
0
0
3
90
Calum Murray retweeted
Everything you need to start self-hosting an open LLM.
Run it on your own hardware. No API keys. No per-token bill. Nothing leaves your machine.
The full path with @vllm_project: batch inference in Python, an OpenAI-compatible API server in one command, and quantized models GB of weights to a quarter of that while keeping 98-100% accuracy.
Walkthrough by @cedricclyburn.
Run it on your own hardware. No API keys. No per-token bill. Nothing leaves your machine.
The full path with @vllm_project: batch inference in Python, an OpenAI-compatible API server in one command, and quantized models GB of weights to a quarter of that while keeping 98-100% accuracy.
Walkthrough by @cedricclyburn.
20
26
132
8,300
Calum Murray retweeted
What is chain-of-thought actually doing inside the model? Our new #EMNLP2026 Findings paper shows reasoning models rebuild the meaning of words in activation space over a single trace, and that these representations causally drive performance. Thread below👇
Our paper "Fluid Reasoning Representations" was accepted to EMNLP 2026 Findings.
The question we asked: when a reasoning model spends 15k tokens thinking, what is actually changing inside it?
Short answer: it is rebuilding the meaning of words in activation space, mid-trace.
1
5
79
6,018
Exciting to see the A2A protocol joining AAIF!
aaif.io/projects/agent…
aaif.io/projects/agent…
0
0
1
26
Just finished upgrading the Kubernetes MCP Server to the 2026-07-28 spec, it was very seamless with the official Go SDK
1
0
2
32
Excited to be joining the first AAIF Ambassador cohort! 138 open source AI community leaders across 41 countries.
As a current CNCF Ambassador as well, will be focusing on CNCF/AAIF collaboration
@AgenticAIFdn @CloudNativeFdn #AAIFAmbassador #CNCFAmbassador
As a current CNCF Ambassador as well, will be focusing on CNCF/AAIF collaboration
@AgenticAIFdn @CloudNativeFdn #AAIFAmbassador #CNCFAmbassador
0
0
2
14