https://github.com/mlop-ai/mlop
Hey HN!I made a completely open sourced alternative to Weights and Biases with (insert cringe) blazingly fast performance (yes we use rust and clickhouse)Weights and Biases is super unperformant, their logger blocks user code... logging should not be blocking, yet they got away with it. We do the right thing by being non blocking.Would love any thoughts / feedbacks / roasts etc
Hey HN!We built mlop (https://mlop.ai), a fully open source (https://github.com/mlop-ai) ML experiment tracking platform, much like Weights & Biases.Unlike the existing competitors we focus heavily on performance (yes our ingestion backend is in Rust), and top tier user experience, and we are fully open sourced with easy self hosting using docker.Why did we build this? WandB was misleading...
I am currently working on an MVP for a MLOps platform much like weights and biases (wandb.ai). I would be very intereted to hear what peoples pain points are with wandb!I will list mine: - Limited collaboration, its hard to share graphs between team memebrs without sc and talking thru slack e.g, I want this to be more refined - Reports, their reports UI is bad and difficult to use - Speed, slow...