Open Source AI Live

Tech Friend AJ · Intermediate ·🤖 AI Agents & Automation ·1w ago

Skills: Agent Foundations90%Tool Use & Function Calling70%

Key Takeaways

Live coding and demonstration of open source AI and real agent workflows using Codex stream setup and DotAgents voice control

Original Description

Live with AJ from TechFren. Today we're digging into open source AI and real agent workflows: Codex stream setup, DotAgents voice control, Hermes Agent, OpenClaw, browser automation benchmarks, Playwright MCP, Wordle as a test harness, and a live Tailscale RTMP clipping pipeline. We move fast across GitHub projects, Codex issues, local model safety, agent control panels, voice agents, repo security checks, vector databases, and live clip infrastructure. TIMESTAMPS 0:00 - Open source intro 3:34 - Stream dashboards 6:56 - Codex set up stream 9:48 - Hermes dashboard workflow 11:18 - Codex browser control 12:25 - OBS CLI and AI careers 18:21 - CrewAI updates 20:07 - Harnesses matter 23:11 - Local model safety 27:59 - Codex hotkey request 29:57 - DotAgents voice demo 33:28 - Codex iOS plugin 34:57 - Terminal bench question 40:32 - Chaos Coder demo 43:56 - Securing env secrets 47:10 - Feature request copy 1:09:51 - Codex token usage 1:13:04 - Alex idea radar 1:17:05 - Speech-to-speech repo 1:28:52 - Agent control panels 1:31:53 - Hermes neural map 1:35:49 - Cmux and Herder 1:41:04 - Canvas voice agents 1:44:07 - OpenClaw vs Hermes 1:48:43 - Importing Hermes skills 1:54:08 - Chrome DevTools angle 1:54:11 - OpenClaw install plan 1:58:12 - Token pricing critique 2:00:53 - OpenClaw voice setup 2:09:12 - Repo safety checks 2:15:39 - ZVec vector database 2:20:17 - Browser agent benchmark 2:28:12 - Playwright MCP bloat 2:30:35 - Wordle bench 2:38:32 - Playwright token costs 2:39:47 - Live clipping idea 2:45:01 - Custom RTMP plan 2:50:10 - Hermes RTMP server 2:52:24 - Sub-agent orchestration 2:57:01 - YouTube poll design 3:14:09 - Live clip dashboard 3:15:17 - Tailscale RTMP ingest 3:16:54 - First live clip 3:21:08 - Clips transcript dashboard 3:23:26 - Agent desktop idea 3:25:11 - Stream wrap-up TechFren: https://techfren.net Subscribe for grounded AI builds, open source AI tools, local model workflows, agent experiments, and practical tech breakdowns.

Watch on YouTube ↗ (saves to browser)

Sign in to unlock AI tutor explanation · ⚡30

More on: Agent Foundations

View skill →

Build and Deploy an Agent with Reasoning Engine in Vertex AI

Adding a Phone Gateway to a Virtual Agent

From Zero to Working AI Agent in 60 Seconds

From Zero to Working AI Agent in 60 Seconds

Create An AI Agent With Replit That Automates Your Sales

Create An AI Agent With Replit That Automates Your Sales

Capstone: Autonomous Runway Detection for IoT

Capstone: Autonomous Runway Detection for IoT

AI Agents with Model Context Protocol & Typescript

AI Agents with Model Context Protocol & Typescript

Related Reads

AI-Powered DevSecOps: How Intelligent Automation Is Transforming Secure Software Delivery

Learn how AI-powered DevSecOps transforms secure software delivery by automating security and compliance checks, and why it matters for modern software development

Medium · Cybersecurity

AI-Powered DevSecOps: How Intelligent Automation Is Transforming Secure Software Delivery

Learn how AI-powered DevSecOps is transforming secure software delivery by automating security testing and compliance, and why it matters for modern software development

Write Your Exceptions Down

Learn to write down exceptions for smoother agent operation and reduced errors

Dev.to · Nabbil Khan

A signature proves who signed the receipt. It does not prove the receipt is true. Here is how you test the difference.

Learn to test the difference between authenticity and validity in signed records using agent receipts

Dev.to · Michael "Mike" K. Saleme

Chapters (46)

Open source intro

3:34 Stream dashboards

6:56 Codex set up stream

9:48 Hermes dashboard workflow

11:18 Codex browser control

12:25 OBS CLI and AI careers

18:21 CrewAI updates

20:07 Harnesses matter

23:11 Local model safety

27:59 Codex hotkey request

29:57 DotAgents voice demo

33:28 Codex iOS plugin

34:57 Terminal bench question

40:32 Chaos Coder demo

43:56 Securing env secrets

47:10 Feature request copy

1:09:51 Codex token usage

1:13:04 Alex idea radar

1:17:05 Speech-to-speech repo

1:28:52 Agent control panels

1:31:53 Hermes neural map

1:35:49 Cmux and Herder

1:41:04 Canvas voice agents

1:44:07 OpenClaw vs Hermes

1:48:43 Importing Hermes skills

1:54:08 Chrome DevTools angle

1:54:11 OpenClaw install plan

1:58:12 Token pricing critique

2:00:53 OpenClaw voice setup

2:09:12 Repo safety checks

2:15:39 ZVec vector database

2:20:17 Browser agent benchmark

2:28:12 Playwright MCP bloat

2:30:35 Wordle bench

2:38:32 Playwright token costs

2:39:47 Live clipping idea

2:45:01 Custom RTMP plan

2:50:10 Hermes RTMP server

2:52:24 Sub-agent orchestration

2:57:01 YouTube poll design

3:14:09 Live clip dashboard

3:15:17 Tailscale RTMP ingest

3:16:54 First live clip

3:21:08 Clips transcript dashboard

3:23:26 Agent desktop idea

3:25:11 Stream wrap-up

AI Agents Explained 🤖

Learn With Shopify