Daniel A.

BSc in Computer Engineering, Politehnica Bucharest. MSc in Artificial Intelligence. Published at ICCV — Inside Knowledge: Graph-based Path Generation with Explainable Data Augmentation and Curriculum Learning for Visual Indoor Navigation.

I work on computer vision, robotics, and the messy bits in between.

• github
• linkedin
• twitter / x
• cv (pdf)
• email

• 2026-05-03 hello, world

  First post.

• LeRobot — embedding-guided grasping

  2026
  robotics computer vision fastsam embeddings

  A LeRobot arm that learns target objects from image embeddings and picks them up. Image embeddings localize the requested object, FastSAM segments it, and the arm plans a grasp from the resulting mask. The pipeline runs end-to-end from a single reference image — no per-object training.