Substack

Autonomous Coding Agents via OpenAI Symphony + Linear

Alessio Fanelli (founder of Kernel Labs) demonstrates running autonomous coding agents from a phone using OpenAI Symphony orchestrated with Linear as an agent state machine. The episode covers a fully autonomous development workflow (Symphony managing agents across the dev lifecycle, Linear providing state tracking), cost and token accounting, and enhanced agent sensing with tools like Glimpse. Fanelli also demos OpenAI Codex autonomously browsing eBay to scout underpriced PSA‑graded Pokémon cards, extracting certificate numbers and flagging $10K–$20K deals for his San Carlos card shop, Merlin Games. The episode is published July 6, 2026 and is available on YouTube, Spotify and Apple Podcasts; sponsors include Firecrawl and Jira Product Discovery.

Guide to Anthropic's Claude Fable 5 and Workflow

A Substack guide by Linas explains practical prompting and operational workflows for Anthropic’s Claude Fable 5. The article notes a limited window — through July 7, 2026 — when Fable 5 is included in Pro, Max, Team and select Enterprise plans for up to 50% of weekly usage; after that Fable sessions will consume paid usage credits. The piece highlights a field guide by Thariq Shihipar from Anthropic’s Claude Code team, details an 8-technique playbook and prompt structure, warns about fallback models (Opus 4.8), pricing math and geopolitical availability risks, and includes a downloadable “Claude Skill” implementing the recommended workflow.

AI Advances: Fable GPU Kernel, Automation, and OSWORLD 2.0

Import AI reports several AI-research developments: Fable wrote a high-performance GPU 'megakernel' that achieved an 18.71x speedup on an RTX PRO 6000 Blackwell versus an optimized PyTorch baseline in KernelBench‑Mega; the Remote Labor Index (RLI) shows frontier models' success at end-to-end online freelance tasks rising from 2.5% (Oct 2025) to 16.1% (July 2026), with Fable 5 scoring 16.1%; OSWORLD 2.0, a long‑horizon benchmark created by multiple universities and organizations, evaluates agents on 108 multi-hour, multi-program computer-use tasks and finds current agents far from reliable (best settings ≈20.6% binary accuracy); and JD published the Oxygen AI Item Center, an industrial-scale LLM/VLM-centric inventory system running at massive scale on Huawei Ascend NPUs. The newsletter frames these items as signals that AI is improving at core R&D, complex computer use, and real-world operational automation, with implications for economic automation and large-scale enterprise systems.