Alibaba's official tech channel, featuring all of its technology innovations.
Quest redesigns long‑running task execution by combining model capability, context management, dynamic reminders, and a minimal Bash‑centric toolset into a closed‑loop Agent architecture that lets AI generate, verify, and deliver complete software artifacts without constant human intervention.
This article explains what Anthropic Skills are, compares them with MCP, and provides a step‑by‑step, best‑practice workflow—including repository setup, OpenSkills commands, naming conventions, and executable skill patterns—to help developers quickly create robust, efficient Skills for Claude and other AI agents.
This article explains how a hierarchical sparse‑attention framework redesigns KVCache storage across GPU, CPU, and remote memory, eliminates bandwidth and capacity bottlenecks, and enables efficient inference for 128K‑token and larger contexts with dramatically reduced GPU memory usage and higher throughput.
DataAgent, built on Spring AI Alibaba, tackles the "last mile" of AI data analysis by combining deterministic workflow orchestration with large‑model reasoning, offering human‑in‑the‑loop feedback, dynamic prompt configuration, hybrid retrieval, containerized Python execution, streaming SSE, multi‑model scheduling, multi‑source connectivity, and secure API‑key management to deliver instant, insight‑rich reports for business users.
The article explains how legacy monitoring based on isolated metrics, traces, and logs cannot keep up with the massive, fragmented, and dynamic data of modern IT systems, and introduces UModel—a graph‑based observability model that bridges data, model, and engineering gaps to enable AI‑driven operations.
This article explains how the rapid evolution of AI agents caused context explosion, why the original fixed‑window memory failed, and how CodeGenius introduced a layered memory system that unloads stale data, deduplicates files, generates structural summaries, and dynamically compresses dialogue to keep prompts stable, reduce token cost, and improve task continuity.
This article details the design and implementation of a lightweight, non‑intrusive data‑collection framework for multiple AI coding tools—covering background challenges, industry solutions, design principles, a three‑layer MCP‑based architecture, concrete hook scripts, telemetry handling, cross‑platform deployment, and future enhancements.
This article explains how to integrate a Human‑In‑The‑Loop (HITL) mechanism into ReactAgent, detailing the motivation, design of interaction, tool description, XML‑based UI rendering, Redis‑driven waiting loop, and the broader architectural parallels with design patterns and other agent frameworks.
This comprehensive guide shows backend developers how to collaborate effectively with AI coding tools by building personal context management systems, accurately judging AI output quality, following a structured PRD‑to‑code workflow, leveraging Python scripts and agent prompts, and applying best‑practice documentation techniques to boost productivity and code reliability.
This article explains the design, challenges, and high‑fidelity architecture of Tair‑KVCache‑HiSim, a simulation tool that models multi‑level KV‑Cache behavior for large‑language‑model inference, predicts latency, throughput and cost under SLO constraints, and validates its predictions against real GPU deployments with sub‑5% error.
