Episode

How I run autonomous coding agents from my phone with OpenAI Symphony + Linear | Alessio Fanelli (Kernel Labs)

Podcast: How I AI
Published: Jul 6, 2026
Duration seconds: 2154
Processing state: processed
Canonical source: https://podcasters.spotify.com/pod/show/pen-name/episodes/How-I-run-autonomous-coding-agents-from-my-phone-with-OpenAI-Symphony--Linear--Alessio-Fanelli-Kernel-Labs-e3lj6bn
Audio: https://anchor.fm/s/1035b1568/podcast/play/122312503/https%3A%2F%2Fd3ctxlq1ktw2nl.cloudfront.net%2Fstaging%2F2026-6-2%2F427261102-44100-2-dce39f6a892e9.mp3
JSON: /v1/public/podcasts/how-i-ai-7304222/episodes/how-i-run-autonomous-coding-agents-from-my-phone-with-openai-symphony-linear-alessio-fanelli-kernel-labs
Markdown: /podcast/how-i-ai-7304222/how-i-run-autonomous-coding-agents-from-my-phone-with-openai-symphony-linear-alessio-fanelli-kernel-labs.md

Actions

POST https://stenobird.com/v1/public/podcasts/how-i-ai-7304222/episodes/how-i-run-autonomous-coding-agents-from-my-phone-with-openai-symphony-linear-alessio-fanelli-kernel-labs/transcription-requests
Idempotently request low-priority transcript generation for this episode.
GET https://stenobird.com/podcast/how-i-ai-7304222/how-i-run-autonomous-coding-agents-from-my-phone-with-openai-symphony-linear-alessio-fanelli-kernel-labs.md
Read the agent-friendly Markdown representation of this episode resource.

Summary

Move from being an AI prompter to an agent manager by using Linear as a state machine for autonomous coding. Learn how to orchestrate complex workflows using OpenAI Symphony and automate niche business tasks like Pokémon card arbitrage.

Topics

OpenAI Symphony
Linear
Autonomous Agents
AI Orchestration
Software Engineering Automation
Agentic Workflows
Web Scraping
Small Business Automation

Highlights

Main idea: Shift your mental model from 'prompting' to 'agent management' to handle complex, multi-step engineering tasks
Practical takeaway: Use Linear as a state machine to track agent progress, acceptance criteria, and historical context
Failure mode: Avoid bloated CLAUDE.md or skill files; frequent purges are necessary to prevent models from following outdated or vague instructions
Practical takeaway: Deploy agents on cloud VPS instead of local Mac Minis to achieve true scalability and 24/7 autonomy
Main idea: AI agents can create entirely new small business categories by automating high-bandwidth, manual tasks like price scouting

Chapters

1:00 The Shift to Autonomous Workflows: An introduction to using AI to reclaim human clock time by automating repetitive, high-bandwidth tasks.
4:00 Prompter vs. Agent Manager: Discussing the evolution from manual prompting to managing mobile-accessible, hosted agent environments.
6:00 Engineering with Symphony and Linear: A deep dive into using Linear for task tracking and Symphony for managing agent workflows and acceptance criteria.
11:00 Scaling Agent Infrastructure: Why moving from local machines to cloud VPS is essential for running persistent, autonomous agents.
14:00 Managing Context and Skill Files: The importance of purging markdown files and maintaining clean, specific instructions to prevent agent drift.
19:00 Automating Niche Arbitrage: How Codex can autonomously browse eBay to identify underpriced Pokémon cards for a specialized card shop.
27:00 The Future of AI-Driven Small Business: Reflecting on how AI creates leverage for small businesses by automating manual data extraction and monitoring.