Machine Learning
GitHub Copilot Gains Cross-Agent Memory System in Public Preview
GitHub launches memory feature for Copilot agents, enabling AI assistants to learn from past interactions and share knowledge across coding, CLI, and code review workflows.
LangChain Unveils Four Multi-Agent Architecture Patterns for AI Development
LangChain releases comprehensive guide to multi-agent AI systems, detailing subagents, skills, handoffs, and router patterns with performance benchmarks.
Multi-Node GPU Training Guide Reveals 72B Model Scaling Secrets
Together.ai details how to train 72B parameter models across 128 GPUs, achieving 45-50% utilization with proper network tuning and fault tolerance.
Selecting the Optimal Open-Source Model for Production Applications
Explore the criteria for choosing the right open-source model for production, balancing quality, cost, and speed, while considering legal and technical factors.
Character.ai Unveils Efficient Techniques for Large-Scale Pretraining
Character.ai reveals innovative methods for optimizing large-scale pretraining, focusing on techniques like Squinch, dynamic clamping, and Gumbel Softmax, to enhance efficiency in AI model training.
Revolutionizing Semiconductor Defect Detection with AI-Powered Models
NVIDIA leverages generative AI and vision foundation models to enhance semiconductor defect classification, addressing limitations of traditional CNNs and improving manufacturing efficiency.
NVIDIA Unveils Nemotron 3: Innovations in AI Model Efficiency and Accuracy
NVIDIA introduces Nemotron 3, an advanced AI model offering enhanced reasoning and efficiency through its hybrid Mamba-Transformer architecture and reinforcement learning capabilities.
Agent Engineering: Bridging the Gap Between Development and Production
Agent engineering is emerging as a crucial discipline in developing reliable AI systems. Learn how it combines product thinking, engineering, and data science for non-deterministic systems.
AutoJudge Revolutionizes LLM Inference with Enhanced Token Processing
AutoJudge introduces a novel method to accelerate large language model inference by optimizing token processing, reducing human annotation needs, and improving processing speed with minimal accuracy loss.
NVIDIA's ToolOrchestra: Revolutionizing AI with Small Orchestration Agents
NVIDIA's ToolOrchestra employs small orchestration agents to optimize AI tasks, achieving superior performance and cost-efficiency. Discover how this innovation is reshaping AI paradigms.