Author

JD Tech Talk

Official JD Tech public account delivering best practices and technology innovation.

390

Articles

Likes

1.2k

Views

Comments

Latest from JD Tech Talk

100 recent articles max

JD Tech Talk

Jan 9, 2026 · Artificial Intelligence

How JoyCode Agent Scored 74.6% Pass@1 on SWE‑bench Verified with a Patch‑Test Co‑generation Loop

JoyCode Agent leverages a patch‑test co‑generation and iterative validation framework to achieve a 74.6% Pass@1 score on the SWE‑bench Verified benchmark, reducing resource consumption by 30‑50% and introducing a closed‑loop multi‑agent pipeline that integrates testing, patch generation, trajectory compression, similarity retrieval, and decision arbitration.

AILLMMulti-agent

0 likes · 41 min read

How JoyCode Agent Scored 74.6% Pass@1 on SWE‑bench Verified with a Patch‑Test Co‑generation Loop

JD Tech Talk

Jan 9, 2026 · Artificial Intelligence

How JoyCode Agent Reached 74.6% Pass@1 on SWE‑bench Verified with a Patch‑Test Co‑Generation Loop

This technical report details JoyCode Agent’s end‑to‑end pipeline that couples patch generation with fail‑to‑pass and pass‑to‑pass test creation, uses trajectory compression, CSR similarity retrieval, and multi‑agent iterative retries to achieve a 74.6% Pass@1 score on the SWE‑bench Verified benchmark while cutting compute costs by 30‑50%.

AI code repairPatch GenerationSWE-Bench

0 likes · 38 min read

How JoyCode Agent Reached 74.6% Pass@1 on SWE‑bench Verified with a Patch‑Test Co‑Generation Loop

JD Tech Talk

Dec 12, 2025 · Big Data

Understanding Hudi Core Concepts: Timeline, Indexes, and Table Types Explained

This article explains Apache Hudi’s core concepts, including its timeline architecture, file layout, indexing mechanisms, and the two primary table types—Copy on Write and Merge on Read—along with their trade‑offs and the various query modes such as snapshot, time‑travel, and incremental queries.

Apache HudiTable Typesbig data

0 likes · 9 min read

Understanding Hudi Core Concepts: Timeline, Indexes, and Table Types Explained

JD Tech Talk

Dec 3, 2025 · Databases

How DongSQL’s Syntax Extensions and Concurrency Controls Supercharge E‑commerce Performance

This article provides an in‑depth technical analysis of DongSQL’s kernel redesign—including RETURNING clause, Inventory Hint, CCL concurrency control, Statement Outline, and thread‑pool optimizations—showing how each feature dramatically improves throughput and latency for high‑traffic e‑commerce workloads.

Concurrency ControlSQLdatabases

0 likes · 12 min read

How DongSQL’s Syntax Extensions and Concurrency Controls Supercharge E‑commerce Performance

JD Tech Talk

Dec 1, 2025 · Artificial Intelligence

How JoyAgent Enables Multimodal RAG for Enterprise Knowledge Management

JoyAgent, JD's open‑source intelligent‑agent platform, now adds multimodal Retrieval‑Augmented Generation (RAG) capabilities, combining graph‑based knowledge, hierarchical chunking, and vision‑language models to handle text, images, tables, and API data for enterprise knowledge processing and evaluation.

Enterprise AIKnowledge GraphMultimodal RAG

0 likes · 11 min read

How JoyAgent Enables Multimodal RAG for Enterprise Knowledge Management

JD Tech Talk

Nov 21, 2025 · Artificial Intelligence

Mastering Chunking Strategies for Retrieval‑Augmented Generation

This article explains why effective chunking is crucial for RAG performance, compares seven major chunking strategies—including fixed‑size, semantic, recursive, document‑structure, agent‑driven, sentence, and paragraph methods—and offers practical guidance on selecting and optimizing chunks for real‑world AI applications.

AIChunkingRAG

0 likes · 10 min read

Mastering Chunking Strategies for Retrieval‑Augmented Generation

JD Tech Talk

Nov 20, 2025 · Artificial Intelligence

Unlocking Heterogeneous Treatment Effects: Theory, Methods, and a CATE Tool

This article explains experimental heterogeneity (HTE), clarifies key concepts such as CATE and ITE, discusses why analyzing treatment‑effect variation matters for business, compares statistical and machine‑learning methods, and introduces an open‑source Python tool that automates CATE discovery and reporting.

CATEExperiment analysisITE

0 likes · 13 min read

Unlocking Heterogeneous Treatment Effects: Theory, Methods, and a CATE Tool

JD Tech Talk

Nov 19, 2025 · Backend Development

Boost Java Microservice Debugging with Easy Code Reader MCP Server

The article explains how Easy Code Reader MCP Server integrates with Joycode and large language models to automatically retrieve and analyze Java source code across multiple projects or JAR files, improving code‑agent accuracy and developer productivity.

AI-assisted developmentJavaMCP

0 likes · 20 min read

Boost Java Microservice Debugging with Easy Code Reader MCP Server

JD Tech Talk

Nov 10, 2025 · Artificial Intelligence

Designing an AI-Powered Experiment Analysis Agent: Architecture, Workflow, and Future Enhancements

This article outlines the motivation, design, architecture, engineering implementation, large‑model selection, and future improvement plans for an AI‑driven experiment analysis agent that integrates data aggregation, modular workflow orchestration, and interactive frontend features to streamline AB‑test insights.

AI AgentExperiment analysisLarge Language Model

0 likes · 14 min read

Designing an AI-Powered Experiment Analysis Agent: Architecture, Workflow, and Future Enhancements

JD Tech Talk

Nov 4, 2025 · Artificial Intelligence

How AI-Powered Virtual Try-On Transforms Fashion E‑Commerce

The article explains how JD.com's AI virtual try‑on system Oxygen Tryon uses advanced computer‑vision and generative models to let shoppers instantly preview clothing on their own photos, dramatically improving purchase decisions, reducing return rates, and outlining technical challenges, innovations, and future development plans.

AIFashion E‑commercecomputer vision

0 likes · 7 min read

How AI-Powered Virtual Try-On Transforms Fashion E‑Commerce