I build AI agents that make decisions for groups of people and show their reasoning.

MeetSpot, for example, tells every person in the group why each venue ranked where it did.

See the work Email me

query: what has Calder actually shipped?

  1. [1] MeetSpot: 535 GitHub stars, 2nd Prize at the Soul App Agent Hackathon Final.github.com/calderbuild/MeetSpot
  2. [2] QualityAgent: first-author paper at IEEE IRAI 2026, plus the IEEE IES SYPA Travel Award./about#research
  3. [3] 19 hackathon builds, 5 of them with a prize or a paper./projects#hackathons
  4. [4] 1.3k+ stars across open-source repos.github.com/calderbuild

4 claims, 4 sources. Checked Oct 2026.

Now

  • AI Engineer Intern at MindLight Tech, remote.
  • Running a pipeline that picks a topic, writes, voices, renders and posts a short video about AI agents twice a day, with nobody approving each one.

Selected work

All projects and hackathons

  • MeetSpot

    A group cannot agree where to meet.

    Picks a meeting point that is fair to everyone and explains why each venue ranked where it did. It measures fairness in real commute time instead of straight-line distance.

    FastAPI / React / Maps APIs

    535 stars

    2nd Prize, Soul App Agent Hackathon Final

    Code Live demo

  • QualityAgent

    Every factory defect has to end up as ISO 9001 paperwork.

    Three agents turn a defect image into CAPA and PFMEA documents that follow ISO 9001, with a human supervising the output.

    Multi-Agent / Vision LLMs / ISO 9001

    First-author paper, IEEE IRAI 2026. IEEE IES SYPA Travel Award.

    Code not public

  • VibeDoc

    A rough product idea is not something you can build from.

    Turns the idea into a PRD, architecture diagrams, and AI coding prompts in a few minutes.

    Gradio / Qwen / Prompt Engineering

    375 stars

    Best Topic Award, OpenManus Hackathon

    Code

  • WeFinance

    Bill photos hold financial data nobody types in.

    Reads bill images with a vision LLM and turns them into spending insights.

    Vision LLMs / Python

    129 stars

    Excellence Award, 12th Nvidia Sky Hackathon

    Code

  • agentcut

    Making a short video takes six different jobs.

    Six agents split scripting, storyboard, voice, visuals, editing and review, and return a finished video from one prompt.

    Multi-Agent / Video Generation

    59 stars

    Code

  • SafeReceipt

    When an AI agent fails, nobody can prove who was responsible.

    Gives each agent an on-chain identity and records its actions as receipts anyone can re-verify.

    Solidity / Monad / Agent Identity

    Contracts deployed on Monad Testnet, public evidence repo

    Code