MODELFRAMEWORK

Apodex 1.1

Agentic intelligence for complex research — combining files, search, code, and agent collaboration.

From reasoning to getting work done.

How can an agent use files, search, code, and collaboration to carry a complex research task through to completion?

My work

Built the deep-research agent harness from scratch at Apodex, with evidence verification and multi-agent coordination.

Outcome

A model and workbench for complex research, alongside the open-source FrontierAgent harness.

Context and my contribution

Apodex 1.1 brings reasoning into research tasks involving files, search, code, and agent collaboration. My work at Apodex included building the deep-research harness and leading its core architecture as the development effort grew to four or five contributors. I designed multi-source checks for research claims and enabled parallel tool use. Since May 2026, my focus has shifted to pretraining data acquisition and quality; the experience page describes these two stages separately.