Intellectual hub of the topic
AI: Events
Technical context • Infrastructure
Red Hat OpenShift AI offers a solution for the rapid processing of unstructured data in RAG systems using a distributed architecture.
Databricks has developed its own approach to creating AI agents – the coSTAR system, which allows the team to work quickly without losing control over quality.
AI: Events
Technical context • Development
The PyTorch team has introduced TorchSpec, a tool designed to facilitate the training of speculative decoding, thereby accelerating the performance of large language models.
AI: Events
Infrastructure
Databricks has introduced AI Runtime – an environment for training and fine-tuning models on NVIDIA GPUs without the need to deploy your own infrastructure.
Cursor has updated its AI coding assistant: the new version runs faster, handles complex tasks better, and uses fewer tokens.
AI: Events
Infrastructure
SK Telecom has successfully had its AI data center interconnect architecture recognized as an international standard, a game-changer for the entire industry.
Researchers have accelerated the training of the Llama 4 Scout model by 30.2% without compromising result quality, thanks to a change in the numerical data format.
At the NVIDIA GTC 2026 conference, Intel announced that its Xeon 6 processors will be used as host CPUs in NVIDIA's DGX Rubin NVL8 systems.
Hugging Face has reviewed the progress of open AI: the growth in models, datasets, and Spaces has hit new highs, while the community continues to expand.
Don’t miss a single experiment!
Subscribe to our Telegram channel —
we regularly post announcements of new books, articles, and interviews.