← back

Research · AI Safety

Checkpoint AI Society

Checkpoint is Korea's first AI prompting research society, focused on understanding and improving the reliability of large language models. As a founding member, I contributed to research on LLM hallucinations, guardrail design, and multi-agent system architectures.

RoleFounding Member
WhenSep 2023 — Feb 2024
StackLLM Research, Prompt Engineering, Multi-Agent Systems, Guardrails

Highlights

  • Co-founded Korea's first AI prompting research society
  • Conducted research on LLM hallucinations, guardrails, and multi-agent system architectures
  • Built and deployed an internal email service used by society members
  • Contributed to the early Korean AI safety research community