The article explains how LLM‑driven agent tool calls differ from traditional function calls in timing, parameter sourcing, error handling, call‑chain observability, and performance, and it provides concrete examples, failure modes, and interview‑ready summaries.
This article explains the core concepts behind AI agents—including how they differ from large language models, their relationship to workflows, the various agent operating modes, and the underlying technologies such as function calls, the Model Context Protocol (MCP), Skills, and the Agent‑to‑Agent (A2A) protocol—providing clear examples and practical comparisons for developers and interviewees.
This article explains the OpenClaw AI agent framework, detailing how its core Agent follows an Observe‑Plan‑Act loop, how Memory uses SQLite for short‑ and long‑term storage, how RAG retrieves external knowledge, and how Skills replace MCP with modular tool workflows, plus security tips and deployment links.
The article explains OpenClaw’s four‑layer architecture—Agent, Memory, RAG, and Skills—detailing how each component (including Function Call and MCP) works, how they differ from platforms like Dify, and provides practical security guidelines for running the open‑source AI agent framework.
The article explains how Text2SQL should be treated as a read‑only tool within an AI Agent, covering its role in function calls, dynamic schema pruning, ambiguity handling, SQL safety checks, result validation, semantic caching, and logging to build a production‑grade system.
This article explains the proper role of Text2SQL within an Agent architecture, detailing its placement as a tool, function‑call implementation, decision logic for invocation, multi‑turn handling, failure management, and how to clearly present these concepts in technical interviews.
This article explains the root causes of hallucinated parameters in LLM Function Calls, outlines five common failure patterns, and presents a systematic five‑step engineering framework—including schema design, prompt rules, dynamic routing, result validation, and clarification—to reliably eliminate such errors in real‑world AI agents.
The article explains why the ReAct (Reason + Act) framework outperforms simple Chain‑of‑Thought prompting by adding executable actions, environment state awareness, and feedback loops, making large language models into controllable, reproducible, and error‑recoverable agents suitable for real‑world applications and interview discussions.
This article breaks down why function‑call reliability is the biggest bottleneck for LLM agents and presents a systematic five‑step loop—schema quality, prompt context, sampling, training data, and runtime defenses—plus concrete optimization techniques such as dynamic tool routing, plan‑execute, validation layers, memory injection, and log‑driven tuning, illustrated with real‑world cases.
System calls differ from regular function calls by using the CPU’s privileged syscall instruction, indirect indexing via registers, and a mode switch from user to kernel space, allowing the OS to control which kernel functions applications can invoke, while ordinary calls use direct addresses and stay in user mode.
This presentation by Yao Yitong of China Telecom AI Research Institute explains why Function Call is essential for LLM deployment, outlines data‑centric and training‑centric optimization methods, discusses common pitfalls and reward‑function design for reinforcement learning, and showcases practical Agent application patterns for real‑world tasks.
On July 12, 2025, the DataFun Summit will feature a technical session where China Telecom AI Research Institute engineer Yao Yitong presents a deep dive into enhancing large language model Function Call abilities through systematic data and training optimizations, offering practical insights for AI practitioners.
This article critically examines the rise of Model Context Protocol (MCP) in AI, explaining its purpose as a unified tool‑calling standard, detailing its architecture, comparing it with traditional function calls, and evaluating the technical and market challenges that limit its universal applicability.
Model Context Protocol (MCP) is an open standard that creates secure bidirectional links between data sources and AI tools, offering ecosystem plugins, cross‑model compatibility, and data‑privacy benefits; the article compares MCP with function calls and agents, outlines its architecture, and demonstrates practical implementations at Huolala.
An in‑depth overview traces the development of AI agents from early LLM milestones to modern “class‑Agent” models, examines core components such as memory, tool use, planning and reflection, analyzes current limitations, and outlines emerging solutions like workflows, multi‑agent systems, and model‑as‑product paradigms.
This article explains how the Model Context Protocol (MCP) decouples tool providers from AI agents, introduces a USB‑C‑like standard for context and tool exchange, and demonstrates its impact on development paradigms, tool ecosystems, and future AI agent architectures.
This article explains the Model Call Protocol (MCP), its purpose, core implementation code, and step-by-step workflow for integrating AI function calls, comparing it with custom solutions, and discussing practical considerations such as accuracy, cost, and ecosystem impact.
The article examines how AI agents differ from static workflows, outlines the ideal characteristics for agent tasks, explores codebase indexing, RAG and Function Call techniques, and introduces the Model Context Protocol (MCP) as a standardized, efficient bridge between large language models and enterprise tooling for next‑generation AI‑driven software development.
This article explains the MCP protocol and Function Call mechanism for large language models, detailing how tools are described, invoked, and processed, and provides practical code examples ranging from OpenAI JSON specifications to fast‑MCP Python and Spring MVC implementations.
This tutorial explains how to overcome the limitations of Retrieval‑Augmented Generation by using the Model Context Protocol (MCP) together with a MongoDB database, providing detailed setup steps, configuration examples, and performance comparisons that demonstrate significantly higher query precision for large language models.
This article explains how to develop robust AI agents by breaking down large prompts, selecting appropriate tools, managing context efficiently, and applying modular design principles to reduce token costs, avoid hallucinations, and improve overall performance and reliability.
This article explains why overloading prompts and tools harms AI‑Agent performance, and offers practical modular design, intent‑driven instruction splitting, and efficient context management strategies such as curated function‑call tools and dynamic RAG to reduce token costs, improve response speed, and avoid hallucinations.
This article introduces the Model Context Protocol (MCP), explains its core features and open‑source implementations, and then compares MCP with Function Call and AI Agent concepts, highlighting their similarities, differences, and practical use‑case examples.
This article explains step by step how Python creates a function object, builds a call stack, handles parameters, executes the body, performs garbage collection, and manages recursion, illustrating each stage with clear code examples and diagrams.
This article explains the technical principles behind Function Call and ReAct in large language models, provides code samples, compares their strengths and limitations, and illustrates each approach with practical scenarios such as smart customer service and financial analysis assistants.
The article explains how operating‑system multitasking and process switching share the same underlying mechanism as function calls, demonstrates a C program that overwrites a return address to jump to an unexpected function, and shows the resulting assembly to illustrate the similarity between buffer‑overflow attacks and legitimate context switches.
This article explores how JD.com's advertising team leverages Agent technology to enhance advertising operations through AI-driven automation, covering technical implementations like RAG, Function Call capabilities, and platform architecture for scalable AI solutions.
This article details JD Advertising's practical use of large‑model agents—including RAG, function calls, and workflow automation—to improve service efficiency, monitoring, and developer productivity across its advertising platform.
JD’s advertising R&D team created the JDZunTong Intelligent Assistant by engineering a modular Agent platform that combines advanced Retrieval‑Augmented Generation (RAG 1.0 → 2.0) and Function‑Call capabilities, a visual designer, custom tool registration, and a native Python workflow engine to deliver intelligent customer service, data queries, and ad creation for merchants.
The article explains the essence of process switching by comparing it to function calls, demonstrates how manipulating a function's return address can redirect execution to another function, and shows the underlying assembly and security implications of such control‑flow changes.
This article explains the low‑level mechanics of Python function calls, distinguishing Python‑implemented and C‑implemented functions, dissecting the bytecode generated for a simple call, and walking through the CPython CALL instruction, stack layout, method handling, and the relationship between PyFunctionObject, PyFrameObject, and PyCodeObject.
This article explains how a function call creates a stack frame that records the return address, registers, local variables, and parameters, detailing the roles of ESP and EBP registers and the step‑by‑step process of saving and restoring data on the call stack.
This article explains the concept of a stack as a LIFO data structure, its hardware implementation in CPUs, how it supports function calls and multitasking, and details the four types of stacks used in Linux—including process, thread, kernel, and interrupt stacks—along with code examples and diagrams.
This article explains CPU registers, stack layout, function call conventions, and provides detailed explanations of how glibc implements memcpy and memset both in C and in low‑level assembly, including code snippets and the handling of remaining bytes.
