Conceptualizing a Hosted Platform for Browser Agents
sarans_31 · reddit · 2026-07-11
The author is seeking feedback on building a browser/computer-use AI agent platform, envisioning it as Vercel is to web apps or Render to servers. Developers would simply create an agent, connect websites/credentials/tools, and click deploy. The platform would handle cloud browsers or VMs, auth and secret management, scheduling, long-running execution, monitoring logs, human-in-the-loop approvals, state persistence, recovery, retries, and scaling.
They want to validate several key questions:
- Would you trust such a managed platform over building your own production infrastructure?
- What are the hardest production challenges for browser/computer-use agents?
- If you use tools like Browserbase, Skyvern, or Playwright, what is missing?
- Would you pay a monthly fee, and what capabilities would make it a "must-have" rather than a "nice-to-have"?
More from coding & agent
- Tenable and AWS launch a Black Hat build event for open-source security agents and MCP servers — Dave_Maynor · 2026-07-22
- Codex helps build Valdiluce, an open-world game with climbing, gliding and gondolas — Dimillian · 2026-07-22
- HeyGen adds a media-sourcing skill for coding agents with 75k images and 10k tracks — HeyGen · 2026-07-22
- Agent search bottlenecks are now about variance, not raw latency — rohanpaul_ai · 2026-07-22
- LangSmith adds tracing for Pipecat, LiveKit, OpenAI Realtime, and Gemini Live — LangChain · 2026-07-22
- An MCP server signs every AI agent tool call into a verifiable Merkle chain — Funky_Chicken_22 · 2026-07-22