CodexCalls is a real-time human-in-the-loop system that enables AI coding agents to safely make high-impact decisions by calling a human when needed. When an AI agent encounters a risky or ambiguous action (e.g. running destructive commands, pushing code, or getting stuck), CodexCalls automatically triggers a phone call. A voice agent explains the situation, captures a spoken decision (approve, deny, or instruct), and converts it into a structured outcome that safely continues or stops execution. This creates a seamless bridge between autonomous AI agents and human oversight without slowing down workflows.
More from Global Codex Hackathon: [Singapore]
AI Tender Assistant is a web app that helps teams decide whether to bid on a tender and then generate a first-pass response package faster. It combines a…
End to end workflow of skills and a custom human annotation app for expert review and skill iteration.

Launchpad is a Codex-powered starter platform for rapidly turning a client brief into a working prototype without leaving the Codex window. It combines…
AutoCTO analyses GitHub organisations using a multi-agent Codex pipeline. It reviews repos for code quality, architecture, and tech debt, then surfaces…
Project Worldview is an AI-powered command center that transforms fragmented project and system signals into a unified, real-time operational…
(+ Mario Attard) Digital Twin is a consent-aware platform that turns interviews, assessments, public footprint, and work signals into an evidence-grounded…