Skip to content
View Qingbolan's full-sized avatar
23 followers · 16 following

Achievements

Achievement: Pull Sharkx3Achievement: Pair Extraordinairex2Achievement: QuickdrawAchievement: YOLO

Achievements

Achievement: Pull Sharkx3Achievement: Pair Extraordinairex2Achievement: QuickdrawAchievement: YOLO

Highlights

  • Pro

Organizations

@Easy-notebook @EasyRemote @Generative-Engine-Marketing @Brain-Cell-Coding @Agent-live

Block or report Qingbolan

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
Qingbolan/README.md

Hi, I'm Silan Hu 👋

PhD Student in Computer Science @ NUS

Research Focus: Data Systems for AI Agents · Agent Infrastructure

Creator of Alive, EasyNet, GEM-Bench, and Easy-Notebook

I believe AI should be accessible to everyone—not just experts.

My goal is to make AI easier to discover, easier to use, easier to manage, easier to organize, easier to protect, easier to govern, and easier to monetize.

Research Interests

I am interested in building system-level foundations for AI agents, especially at the intersection of:

  • Database systems for AI agents
  • Hierarchical & dynamic knowledge structures
  • Agent behavior versioning, reuse, and evolution
  • Agent execution infrastructure & distributed systems

My current research direction focuses on designing AI-native data systems that support:

  • Open-World Capability Routing under Capability Drift and Partial Observability
  • evolving agent knowledge,
  • traceable reasoning paths,
  • and reusable agent behaviors as first-class system assets.

Selected Projects

EasyRemote / EasyNet

A distributed execution infrastructure for AI agents and functions.Designed to support language-agnostic agent behaviors, privacy-first compute sharing, and agent-level orchestration.

Python · Go · gRPC · Distributed Systems

GEM-Bench

A benchmark and evaluation toolkit for ad-injected LLM responses. Designed to study the satisfaction–engagement trade-off across chatbots and AI Overviews, with curated datasets, human-validated judges, and reproducible CLI/agent workflows.

Evaluation · Benchmarking · Agent Alignment

Technical Stack

  • Languages: Python, Go, Rust, TypeScript
  • Systems: Databases, Distributed Systems, Agent Infrastructure
  • AI: LLM Agents, Planning, Reinforcement Learning
  • Tools: React, Tauri, gRPC, Docker

Contact

I believe AI agents will become long-running system entities, and we need new database and infrastructure abstractions to support them.

Pinned Loading

  1. EasyRemote/EasyRemote EasyRemote/EasyRemote Public

    🚀 "Torchrun for the World" - Execute local functions on global computing resources while keeping data private. Next-generation distributed computing with zero cold-start latency.

    Python 4

  2. Generative-Engine-Marketing/GEM-Bench Generative-Engine-Marketing/GEM-Bench Public

    First complete benchmark for Generative Engine Marketing (GEM), an emerging field that focuses on monetizing generative AI by seamlessly integrating advertisements into Large Language Model (LLM) r…

    Python 18 1

  3. Silan-Viking Silan-Viking Public

    Make Research Work Findable.A local-first research publishing workspace that helps researchers build an AI-native public presence by keeping every research asset current, evidence-backed, and finda…

    TypeScript 4

  4. Print-SoC Print-SoC Public

    A lightweight web interface for NUS SoC students to submit print jobs securely through SSH, with built-in booklet page arrangement and PDF layout optimization.

    JavaScript