7487 (BingEdward) · GitHub
Skip to content
View 7487's full-sized avatar
🎯
Focusing
🎯
Focusing
15 followers · 36 following

Achievements

Achievement: Pair Extraordinairex3Achievement: Pull Sharkx3Achievement: Quickdraw

Achievements

Achievement: Pair Extraordinairex3Achievement: Pull Sharkx3Achievement: Quickdraw

Block or report 7487

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
7487/README.md

简体中文

BingEdward

AI Agent & LLM Engineering

Building reliable AI applications and the systems behind them.

My work spans AI coding tools, LLM applications, and backend engineering. My background includes cloud / IaaS and GPU-related development at Baidu, and AI coding tools at Alibaba Taotian.

Focus: Agent workflows · AI Code Review · SDK reliability · CI/CD · Retrieval & memory

Selected upstream contributions

A selection of merged pull requests, focused on correctness, compatibility, and regression prevention.

Project My contribution Merged PRs
AMD GAIA Fixed a release-test gate that silently swallowed failures; corrected backend-test dependencies and added workflow contract tests. #3573
PR-Agent Fixed inline-comment line resolution in the Bitbucket Cloud and Server providers, preserving the existing integer-line path. #3136
OpenSandbox Bounded readiness-polling sleeps by the remaining timeout budget in the Go, JavaScript, and Kotlin SDKs; added regression coverage. Go #1744 · JS #1745 · Kotlin #1747
Semantica Removed redundant vector deletion during memory erasure; removed signed Neptune authentication material from logs and added regression tests. Memory #1379 · Security #1373
Apache Superset Isolated table-cell formatting errors so a problematic value can fall back safely without disrupting other cells. #39658

Explore more merged contributions →

Personal tooling

Harness Loop Coding

Workspace templates for Claude Code and Codex CLI, with separate layouts for agent instructions, MCP setup, and shell-driven coding loops.

The repository keeps the two toolchains separate: cc/ for Claude Code and codex/ for Codex CLI.

How I approach engineering

Reproduce the failure → understand the contract → make a focused change → add regression coverage.

I care about explicit failure handling, predictable timeout behavior, backward compatibility, and tests that catch the original bug—not just exercise the happy path.

Working stack: Python · Go · Java · TypeScript / JavaScript · GitHub Actions

Open to technical discussions about agent tooling, AI-assisted development, and reliable LLM systems.

Starred repositories

Browse my categorized GitHub Stars (简体中文) · JSON index

Pinned Loading

  1. Harness_Loop_Coding Harness_Loop_Coding Public

    Shell

  2. amd/gaia amd/gaia Public

    Build AI agents for your PC

    Python 1.6k 179

  3. The-PR-Agent/pr-agent The-PR-Agent/pr-agent Public

    🚀 PR Agent: The Original Open-Source PR Reviewer. This project is not the Qodo free tier.

    Python 13.5k 2k

  4. opensandbox-group/OpenSandbox opensandbox-group/OpenSandbox Public

    Secure, Fast, and Extensible Sandbox runtime for AI agents.

    Python 16k 1.5k

  5. semantica-agi/semantica semantica-agi/semantica Public

    Graph-Native Infrastructure for Context and Accountable AI Systems

    Python 14.2k 1.7k

  6. apache/superset apache/superset Public

    Apache Superset is a Data Visualization and Data Exploration Platform

    Python 75.2k 18.5k