---
date: 2026-04-10
---

# 2026-04-10

### Morning — Before lab

1. **Design end effector** — get the CAD done before heading to campus so it can print during lab meeting
2. **Call DMV** — do it in the car on the way in. No more delays.

### End of day update

Good day. Lab meeting, good lunch with friends, saw cousin on campus — solid social balance.

**Robot arm progress:** Designed and printed the last 3 links. Just missing the end effector (almost done printing). Ahead of schedule on the Tuesday deadline.

**Tomorrow (Fri):** Come back after volleyball, wire up the arm and calibrate it.

**DMV:** Did not call. Try tomorrow (Fri) if they take phone calls, otherwise Monday. No more slipping.

### Research note — cross-embodiment training

Cross-embodiment training is worth exploring for the PARA paper. Could strengthen the generalization story alongside viewpoint and object position robustness.

### Evening — feeling overbooked

Dinner at aunt's house, then bowling with Izzy and her friends. Feeling overbooked and needing personal space. Recurring tension: Izzy likes to pack the social calendar, I need breathing room. It's okay tonight — meditating in the car, 35-40 min at aunt's then bowling. But in general, don't like being rushed like this. Worth addressing the pattern, not just tolerating it.

### Robot build — smaller double-jointed arm

New idea: build a smaller-than-SO100 arm that's double-jointed. Goal is rapid prototyping — smaller means faster prints, cheaper iteration, quicker design cycles.

**Weekend plan (Sat + Sun):** Build the smallest possible robot arm first to build confidence. Then make the double-jointed version. Then scale up link lengths. Progressive complexity, not all at once.