AI Agents
AI Agent technology and applications
Articles
Building a Policy-Governed Multi-Agent Financial Research⭐10
Learn how to build a policy-governed multi-agent financial research workflow using Omnigent, featuring live FX data, delegation, and AI governance.
Google DeepMind Ships Three Physical AI Models for Whole-Body⭐7
Google DeepMind releases Gemini Robotics 2 with three AI models enabling whole-body control, dexterity, and multi-robot collaboration.
Tencent Open-Sources AngelSpec: A Unified Training Framework for⭐8
Tencent open-sources AngelSpec, a unified training framework for speculative decoding draft models, supporting MTP and block-parallel DFlash architectures
Meet Token Saver: An Open-Source MCP Extension Using Local⭐8
Token Saver: an open-source MCP extension using local hybrid RAG to cut Claude PDF token costs by 90-99%.
Moonshot AI Open-Sources MoonEP: A Perfectly Balanced Expert⭐8
Moonshot AI open-sources MoonEP, a balanced expert parallelism library optimizing Mixture-of-Experts model training with enhanced efficiency and scalability.
Prompt Engineering vs Loop Engineering vs Graph Engineering:⭐8
Prompt, loop, and graph engineering are distinct layers of AI control: one response, one agent cycle, and multi-agent orchestration, each building on the last.
Prompt Engineering vs Loop Engineering vs Graph Engineering:⭐8
Three layers of AI control: Prompt, Loop, and Graph engineering. Each layer builds on the last, from single responses to multi-agent orchestration.
Liquid AI Releases LFM2.5-Encoder-230M and LFM2.5-Encoder-350M:⭐7
Liquid AI launches LFM2.5-Encoder-230M and 350M bidirectional models optimized for fast 8K context inference on CPU, enabling efficient real-time edge NLP.
Fireworks AI Launches Fireworks Nexus: A Drop-In Routing and⭐7
Fireworks AI launches Fireworks Nexus, a routing and cost-control layer that shifts routine coding tasks to open-weight models, cutting AI service costs by 30–5...
Building Non-Interactive Agentic Coding Workflows with Moonshot⭐8
Build non-interactive AI coding workflows using Moonshot Kimi CLI, JSONL streaming, testing, and session memory for automated dev pipelines.