Rudel has been the missing “stats layer” for how I actually use AI to write code. Instead of guessing what helps or wastes tokens, I finally see real patterns: which prompts work, where AI speeds me up, and where I still need to step in. The fact that it’s free, open source, and self‑hostable makes it feel like a tool built for engineers first, not just another SaaS dashboard.
What needs improvement
If I had to nitpick, I’d love to see even deeper insights over time (e.g. team‑level trends, more opinionated “here’s what to change next” suggestions) and maybe tighter integrations with other dev tools I already use. But even today, it already feels like one of the best options if you care about understanding how you and your team actually code with AI in the real world.
vs Alternatives
it’s laser‑focused on real developer behavior (sessions, prompts, token spend, where AI helps vs. slows you down) instead of vanity metrics. Being open source and self‑hostable is a huge plus too, since I don’t have to ship sensitive code or prompts to yet another black‑box SaaS.
hey vafa, thank you so much for your review. really love the new features you propsed. especially the opinionated suggestions. we will work our ass off to make that thing even better. thanks for your trust
Opaline
hey vafa, thank you so much for your review. really love the new features you propsed. especially the opinionated suggestions. we will work our ass off to make that thing even better. thanks for your trust
best,
evren