Open Model Strategy & Training
Model Strategy & Training
Track open and local model strategy across Qwen, Gemma, Kimi, DeepSeek, model routing, fine-tuning limits, and training economics.
View Research →Trilogy AI Center of Excellence
Practical AI research, implementation notes, and field-tested resources for leaders and builders working AI-first.
Substack
Get the AI COE signal as it ships
New research briefs, technical walkthroughs, and practical AI implementation notes from the Trilogy AI Center of Excellence.
Open SubstackFeatured Research
Highlights from our ongoing and completed research initiatives.
Autonomous Coding Agents
Agentic Engineering
Investigating AI-driven software development with agent harnesses, coding assistants, review loops, and engineering workflows that can operate across real repositories.
View Research →Agent Federation & Protocols
Agent Infrastructure
Explore open protocols for AI interoperability (A2A, MCP, OGP) to enable reliable task handoff, tool access, and gateway federation across agents.
View Research →Latest Publications
Stay updated with our newest findings, analyses, and insights.
How to Unlock 1M Context in Codex
Agentic Engineering | 2026-08-20
A practical guide to enabling one-million-token context windows in OpenAI Codex using custom model catalogs and server-side configuration.
Read Publication →Grok 4.6, GLM-5.3, Qwen3.8-27B in a 7-Model Week
Model Strategy & Training | 2026-08-16
Seven AI models landed between August 10 and 15, from Grok 4.6 entering Cursor and Grok Build to Qwen’s 2.4T checkpoint setting the scale ceiling.
Read Publication →Codex Chats Disappear when Switching Providers
Agentic Engineering | 2026-08-06
A SQLite bridge script restores missing Codex conversations after switching model providers, by rewriting the provider label that filters the session list.
Read Publication →