all things ai and coding while streaming, DM for consulting.
Apr 8 ⢠6 tweets ⢠3 min read
launching `data-label-factory` a generic auto-labeling pipeline. You write a YAML for your object class and run one command: a vision dataset on a 16 GB MacBook. No GPU, no labelers, no vendor. Using Gemma 4 @Google mlx-vlm @Prince_Canuma + Falcon Perception @lkhphuc
here's how : š§µ(point your /agent at the claude.md file to start asap)
your object class and run one command:
data_label_factory pipeline --project projects/drones.yaml
Stages: gather images from DDG/Wikimedia/YouTube ā VLM filter for yes/no ā
Falcon Perception draws bboxes ā second VLM verifies each box ā review what's left in a browser. All of it runs locally on Apple Silicon.