Viewing profile — manos-saratsis

Tool from tool differs, we do a check of intent vs actual changes https://dromeas.ai/code-review

Very true. A few things that worked for us: 1. Split review priority by blast radius, not diff size. A 400-line PR touching one isolated module is often faster to review than a 40-…

We analysed more than 600 open source repositories and amongst others we see 63x more lines per PR than per direct commit. When teams used to review a handful of changes to stay up…

Spot on and nice complement to the pre-merge half of this problem. Silent logic bugs slips through when a diff "looks right" — what you're building catches them once they're live.

An AI-assisted team. There are many ways to prevent issues from the dev laptop, to team code review, to pre-build test, to automated agentic testing, monitoring, canary deployments…

Very true Sylvain. The evolution of the code review agents is on have configurable autonomy, work on the human learning side, prevent issues both on pull requests or trunk based de…

Manually yes, PRs are often huge lately making it impossible to go through them. Code review bots help going through them. Some interesting stats here https://dromeas.ai/blog/pr-vs…

With today challenges with code reviews here our point of view: multi-model code review with built-in cross-validation for security and compliance findings avoiding single-model bl…

if you wanna some ai generated documentation for OpenClaw here you go https://orchestrai.dev/documentation/openclaw-4a80f8/user_facing/getting_started/73a6987b-6acc-46b2-b8e5-f0c86…

I'm Manos, co-founder of Micro Culture. As a remote-hybrid digital worker I personally had challenges with chat communication. We often chat when we are stressed, when distracted d…