Swati Rawat

Really enjoyed this! The point that agentic AI needs boundaries, verification and the right level of autonomy really resonated with me. But I’m curious about one thing: as we add stronger testing, review and rollback loops around agents, do you think the 40% failure rate would drop significantly, or are we simply moving the failure point elsewhere?

Really enjoyed this perspective. I especially liked the emphasis on boundaries as it is easy to focus on how much faster AI can help us write code, while overlooking the importance of knowing where it shouldn’t be making decisions for us.