Optic: Your final next Agent Infra, built for individuals.

Hello everyone! I’m an indie developer. After trying desktop LLM clients like Claude, ChatGPT, and Codex, I decided to build my own. The LLM clients currently on the market have some flaws I can’t accept: - They’re either tightly tied to specific brands - Or their features are too basic to meet my needs - Or they’re overly bloated, with a steep learning curve and clunky usability Plus, my life isn’t just about one thing. I love programming, I use Obsidian to learn, Notion to manage personal projects, and so on. I don’t want to be constantly switching between Claude Code and various “Copilot” extensions. That’s why I developed Optic. It features a user-friendly interface and powerful Agent capabilities that I believe are more than capable of handling most assistive tasks. It works great as a vibe coding tool and is equally at home in any general-purpose folder. Optic addresses many common needs. For example, you can continue monitoring your desktop work from a mobile browser (just pair devices—no login or registration required), start multiple conversations simultaneously, and more. Optic is highly advanced in terms of LLM infrastructure. We include a reliable context compression mechanism (it won’t compromise instruction following without your knowledge—everything is under control), and we support various advanced features such as code execution, Tool Search, and post-processing of web-fetched results. In the future, we’ll also introduce powerful Subagent capabilities, significantly enhancing Optic’s capabilities. Currently, we’re running a 28% off for launch promotion, and you can get an additional 10% off by using the promo code PRODUCTHUNT. We hope you’ll support us! If you have any questions or feedback, please don’t hesitate to reach out!