HomeCompaniesCompyle

Lovable for Software Engineers

Compyle is a less autonomous coding agent that asks before it acts. We're like Cursor or Claude Code - but you won't need to rewrite your codebase every time you use it for more than a small task. We built Compyle because we were frustrated with the direction other coding agents are headed. They're all racing to be more autonomous, running for hours without human input. This doesn't work for real production code. When agents run autonomously, they make countless small decisions that pile up in your codebase. Engineers lose ownership - they don't know what was implemented, can't effectively debug or maintain it, and struggle to build on top of it. The code becomes a black box, and they're forced to throw it away. Compyle keeps engineers in the driver's seat.
Active Founders
Jonathan Miranda
Jonathan Miranda
Founder
Jonathan is co-founder and CEO of Compyle. He studied Computer Science and Economics at Johns Hopkins University, where he led multiple AI research projects for the CS department and developed snake robots. He also founded Hopkins Deliveries, a campus food delivery service that built cost-effective tele-operated delivery robots. Before Compyle, Jonathan was founding engineer #4 at Hadrius (YC W23), where he built core AI systems powering compliance at thousands of businesses today.
Mark Nazzaro
Mark Nazzaro
Founder
Mark is the Co-founder and CTO of Compyle. Previously the second engineer at Hadrius, he helped scale the company to its Series A, led engineers, and built much of the core AI product. Before that, at arXiv, he led the move to the cloud and launched its first major UI change in decades.
Company Launches
Compyle - Lovable for Software Engineers
See original launch post

TL;DR

Compyle is the coding agent that actually collaborates. It doesn’t just guess what you mean - it asks, plans, and checks in before writing a single line of code. That means you can trust it with bigger and more ambiguous projects than any other coding agent.

https://youtu.be/iNQ8G4k03gc

🙏 Our ask

Try our beta today at compyle.ai and let us know what you ship!

Just connect your Github, select your repos, and create a task to get started.

The Problem

Today’s coding agents - like Codex, Claude Code, and Devin - excel at small, well-defined tasks. Their autonomous design works best for jobs that need little clarification, such as updating libraries or making simple UI edits.

But full autonomy becomes a weakness for open-ended work like building new features or products. These agents often press ahead despite unclear or conflicting requirements - and even when they succeed, the resulting code may be hard for teams to understand.

Good engineers know to clarify requirements before coding, check in with product owners to confirm goals, and collaborate with peers to ensure quality. Coding agents should do the same.

🚀 How Compyle solves this

Compyle keeps you in full control through a question-driven approach. You describe the task, then Compyle asks questions until it understands what you want to build. It produces research and planning artifacts before writing any code.

During implementation, Compyle validates its changes against your decisions. If something doesn't match up, it stops and asks rather than pushing forward. You stay in the driver's seat the entire time.

Our story

Before Compyle, we were building SmartAppetite (AI agents for insurance). Using tools like Claude Code and Cursor, we had a demo in days - then spent weeks cleaning up a codebase we no longer fully understood.

We had to pause feature development just to refactor. That pain is why we built Compyle.

Compyle
Founded:2025
Batch:Fall 2025
Team Size:2
Status:
Active
Location:San Francisco
Primary Partner:Andrew Miklas