Video Library
Workshops, demos, and deep-dives on building and running autonomous agents with Trinity. Grouped by topic, newest first within each group.
Product Demos
Short narrated films, one capability each, recorded on a running Trinity 1.0 instance.
A home for your agents
1:25Not another chatbot
1:18Yours, with receipts
1:27The company brain
1:19Reach is drawn, not inherited
1:40Institutional memory that answers
1:48Agents address roles; the platform finds the person
1:26See what the company knows
1:23Measured against the same numbers people are
1:36Every answer names the version of the facts it used
3:23Delegation with receipts
1:36Who may approve what is company policy, not a setting
2:08Anything irreversible waits for a person
2:09How to require approval for a skill
1:13How to create an agent from the Library
1:02How to give an agent its credentials
1:11How to talk to an agent from its Chat tab
1:03How to finish first-time setup
1:21How to connect Claude, GitHub, email and Gemini keys
1:15How to put an agent on a schedule
1:24How to connect an agent to Telegram
1:11How to find your way around Trinity
1:15How to work with your agents in the Workspace
1:20One Inbox for everything that needs you
1:32A mistake is paid for once
2:47Nothing an agent does is off the ledger
1:35EnterpriseA company, not a chatbot
1:06The brief arrives without being asked
1:17A project with people and agents in it
1:26EnterpriseAn agent is a repository you own
1:37An agent has a role, an objective and people
1:12An agent gets more capable only when a person says so
2:15Drift between teams is caught by the system
2:10One screen for working with every agent
1:47Start Here — Platform Overview
Inside Trinity: Agentic Operating System for AI-Native Companies
The plain-language overview, interview style (24 min): what Trinity is, how a 10-person company runs on a fleet of agents, owning your intelligence, security, the human role, and how to try it
Trinity 1.0 — The Complete Course
Everything in Trinity 1.0, end to end, in one 76-minute video; jump to any of its 41 chapters from the YouTube timeline
Trinity v0.9.0 — Release Tour
Multi-agent org structures, Workspace in every build, the shared skills library, bidirectional A2A
Trinity Architecture Explained
Animated explainer (4:44): one scheduled task end to end — container, gate, delegation, approval, audit, breakers
Trinity v0.8.0 — Release Tour
Voice replies across channels, Brain Orb, Grid dashboard, enterprise identity
Trinity Platform Demo
Full UI walkthrough, end to end
The Multi-Agent Platform I Run My Company On
Live fleet tour: delegation, memory, observability
The Sovereign AI Setup That Changed Everything
The Platform That Runs My Entire Business
Trinity: Secure Infrastructure for Autonomous AI Agents at Scale
Why I Stopped Renting AI Agents (And Built My Own)
The sovereign-vs-SaaS case
Trinity: Sovereign Infrastructure for Autonomous AI Agents
Workshops — Live Recordings
Giving My AI Agency Payment Rails and a Finance Agent (Agent-Native Agency Workshop, Session 6)
Session 6 of six, the finale: what a week of unattended schedules taught us, Agent Builder gating its own deploy behind an approval, one credential vault per agency with a read-only service account, Stripe on a restricted API key, and a finance agent for invoicing, payment recording and accounts receivable
I Let an Agent Make Every Decision on My New Website (Agent-Native Agency Workshop, Session 5)
Session 5 of six: the seats / roles / people / agents model, approvals that route to seats in the canon, a website agent built live from the Trinity template, and a website form flowing to a GitHub issue, the orchestrator, the sales agent and a one-click approval in the rep's inbox
The Moment My AI Agents Talked Me Out of a Discount (Agent-Native Agency Workshop, Session 4)
Session 4 of six: the orchestrator and the Canon — the routing map of who owns what kind of work, a live hand-off to the right department agent, an escalation landing in a human inbox for approval, then both built by hand in the Keymaker agency; the takeaway: restrict agents to prescribed fallbacks
The Open-Source Second Brain for Your Company (Agent-Native Agency Workshop, Session 3)
Session 3 of six: Cornelius as the agency's second brain — the vocabulary (atomic notes, insights, lenses, provenance, the endorsement gate, canon), installing it from a template with first-run indexing, how knowledge gets in and how it is retrieved, and agents working with the brain live
Staffing an Entire Agency With AI Agents (Agent-Native Agency Workshop, Session 2)
Session 2 of six: staffing the Keymaker Agency — sales, marketing, and delivery agent templates from the library, a live sales-agent run (including a correct refusal), a Claude Code agent converted with one plugin command and deployed from GitHub, then permissions so the four agents can talk — plus Q&A on open source vs enterprise, model routing, and the 1.0 roadmap
40 Agents, One Instance (Agent-Native Agency Workshop, Session 1)
Session 1 of six: inside the production instance behind ~40 agents, the v0.9.5 Workspace, shadow agents vs. shared ones, then a DigitalOcean install from zero — plus auditing and migrating the agents you already run
Make Your Agency AI Native (Webinar)
Opening session of the Agent-Native Agency series: inside a production fleet of ~40 agents as an org chart, and how to stop agents sitting in silos
AI-native Agency — Practical Workshops on Trinity
Invitation to the Series of Six: how agents across support, research, ops, and knowledge share context and compound what the company knows
How I Run My Company with a Fleet of 30 AI Agents
Hierarchical fleet as an org chart — orchestrator, managers, workers; shared canon, permission matrix, human gates
Make Claude Code Agents Run Autonomously on Trinity
Analyze, standardize, and migrate home-grown agents into a managed Trinity fleet with the agent-dev plugin
How to Build a Second Brain for Your AI Agents
The 5 rules of agent knowledge: Decisions / Facts / Judgment, knowledge as code, human-endorsed canon
Building & Deploying Agents
Staffing an Entire Agency With AI Agents (Agent-Native Agency Workshop, Session 2)
Install agent templates from the Trinity library, convert an existing Claude Code agent with one plugin command, deploy it straight from a GitHub repository, and wire inter-agent permissions — four agents live on one instance
Make Claude Code Agents Run Autonomously on Trinity
Analyze, standardize, and migrate home-grown agents into a managed Trinity fleet with the agent-dev plugin; deploy from GitHub
One Agent Per Client: A Scalable AI Service Platform
The one-agent-per-client pattern for agencies — sovereign, isolated agents with their own data, credentials, channels, and real guardrails
Build an AI Recruiter Agent with Trinity + Claude Code
Zero to deployed, channel-exposed agent
I Built a DevOps Agent That Deploys Other Agents
Provision and self-host a Trinity instance
Why Every AI Agent Needs a GitHub Repo
Git-as-state, backlog work loops, dashboards
3 AI Agents Run My Entire Software Development Pipeline
The dev-methodology SDLC
Plugins + Trinity: Build and Deploy Agents in Cursor
No terminal required
Building Agents — Playbooks, Plugins, Deployment
From Zero to Deployed AI Agent
Your first Trinity agent
My Dev Workflow for a Large System Using Claude Code
Automation & Loops
Build Autonomous Loops for Your AI Agents
Loops vs. schedules, stop signals, failure modes
Roles, Access & Governance
Trinity Roles & Permissions Explained: Platform Roles, Agent Access, and Approval Relationships
The three permission layers (13 min): platform roles, access to one agent, and who an agent works for and waits on — plus the three rules people get wrong
Multi-Agent Orchestration
The Moment My AI Agents Talked Me Out of a Discount (Agent-Native Agency Workshop, Session 4)
An orchestrator that routes work, tracks receipts and handles escalations, plus a shared Canon with the routing map and approval chains — live on ~40 agents, then built from scratch
40 Agents, One Instance (Agent-Native Agency Workshop, Session 1)
The three pieces that make a fleet behave like one organization — Canon, Cornelius, and the Orchestrator — shown on a live 40-agent instance
How I Run My Company with a Fleet of 30 AI Agents
Hierarchical fleet as an org chart — orchestrator, managers, workers; shared canon, permission matrix, GitHub-issue PM, human gates
How We Automate Our Own Ops & Marketing
A real multi-agent business system, live
Building AI Agent Swarms That Actually Scale (Mrinal Wadhwa)
Architecture deep-dive
Memory & Knowledge Agents (Cornelius)
The Open-Source Second Brain for Your Company (Agent-Native Agency Workshop, Session 3)
Install Cornelius from a template inside your own Trinity, run first-run indexing, and see how it ingests, connects and retrieves knowledge for a fleet of agents
How to Build a Second Brain for Your AI Agents
The 5 rules of agent knowledge: Decisions / Facts / Judgment, knowledge as code, a mind-center agent, human-endorsed canon
Cornelius AI Agent — Architecture Explained
Insights → typed knowledge graph → autonomous reasoning to convergence, gated by human endorsement
Cornelius & the Brain Orb
A second-brain agent renders and talks to its own 3D knowledge graph
Ananda — a Second Brain Built on Cornelius
Another knowledge agent built on the Cornelius template and Brain Orb
I Gave My AI Agent an Unconscious Mind
Autonomous perception → deliberation loops
I Let Cornelius Analyze the Hantavirus Outbreak
Structured analytical reasoning loop
I Gave My AI Three Years of My Notes. Then I Interviewed It.
Voice chat with a knowledge agent
I Gave My AI Agent a Brain — Here's What Happened
Memory as a shared, evolving service
Second Brain AI Agent with Swappable Brains
AI Second Brain with Claude Code and Obsidian
The original build
Use Cases & Case Studies
Measurably Self-Improving Multi-Agent System Without Retraining
Brier HQ — a forecasting fleet with five feedback loops, no retraining; walkthrough of the closed-loop-forecasting research
How I Built an Agent That Makes Movies While I Sleep
Long-running creative production + heartbeat
Investigative Analysis of Epstein Files Using an AI Agent
Knowledge graph over a messy corpus
Control My DGX Spark From Anywhere
Trinity on local/edge hardware over Tailscale
I Let AI Agents Build My Entire Webinar
Agent as a daily-work collaborator
How AI Agents Power Our B2B Marketing Pipeline
From Transcript to Scheduled Posts in 30 Minutes
The content pipeline, short cut
My Secret Content Creation Pipeline: Video to Multi-Platform Posts
Concepts & Strategy
How to Implement AI Support Without Risking Customer Trust
Phased, safe rollout methodology