Skip to content
View hjqcan's full-sized avatar

Block or report hjqcan

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.

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
hjqcan/README.md

hjqcan · swordlight_ai

I build infrastructure for intelligent systems that must keep working after the demo.

REMEMBER → REASON → COORDINATE → EXECUTE → VERIFY

I work where AI meets real systems: persistent memory, multi-agent orchestration, robotics, industrial automation, and verifiable execution.

我关心的不只是“让模型再多答一道题”,而是让智能系统在真实世界中长期运行: 记得住、协同好、执行稳、结果可验证。

Selected systems

System concern Project What it does
Remember GoodMemory A durable, auditable memory layer for AI products and coding agents, with Codex / Claude Code integration, MCP access, local-first storage, and evidence-gated benchmark claims.
Reason together Tachikoma A research-alpha TypeScript multi-agent coding runtime exploring orchestrator-worker collaboration, authoritative task state, checkpoint/resume, MCP, isolated execution, and observability.
Coordinate SwarmRoute MAPF A .NET multi-robot path-planning simulator and reference engine using space-time reservations, liveness policies, SIPP, explicit non-convergence, and visual replay—not a production FMS.
Execute OpenLineOps An early-stage automation-line IDE/runtime exploring immutable release artifacts, device integration, flow orchestration, plugins, production execution, and traceability.
Verify ProofAlpha A paper-first prediction-market automation system for AI-assisted strategy discovery, self-improvement, risk-gated execution, audit evidence, and Arc contract settlement.

The through-line

Across these projects, the recurring problem is trust:

  • memory should be inspectable, correctable, and deletable;
  • agent execution should be isolated and observable;
  • robot coordination should be reproducible and fail honestly;
  • production should run from immutable releases with traceable evidence;
  • autonomous strategies should be paper-first, gated, and auditable.

evidence over demos · explicit boundaries over hidden magic · safe defaults before autonomy

Toolbox

C# / .NET · TypeScript / Bun · React / Electron · Python · PostgreSQL · Solidity

MCP · DDD / Clean Architecture · event-driven systems · agent evaluation · MAPF / SIPP


X / @swordlight_ai  ·  hjqcan@163.com  ·  All repositories

Pinned Loading

  1. GoodMemory GoodMemory Public

    Local-first, auditable memory layer for AI apps and coding agents — Codex, Claude Code, MCP, HTTP, TypeScript, and Python.

    TypeScript 7 1

  2. ProofAlpha ProofAlpha Public

    Verifiable AI-generated prediction market strategies, settled on Arc.

    C# 1