Finally got a 2‑qubit entanglement on my cheap rig

Spent the last week wrestling with the control pulses on my $500 superconducting board and, surprise, the Bell state finally showed up on the oscilloscope. The calibration script was a mess of hard‑coded offsets that I kept tweaking until the fidelity crept past 78%. It felt like trying to tune a piano with a hammer—painful, noisy, and somehow satisfying when a single note finally rang true.

What really got me was the error mitigation step. I threw out the standard readout correction and replaced it with a simple linear regression on the raw counts. The result? A clean violation of the CHSH inequality without any post‑selection. No magic, just a stubborn refusal to accept the default settings.

Anyone else still fighting the same low‑level noise floor, or am I just lucky? Expecting the community to point out the obvious flaws now—because I’m sure there’s a hidden bug in my timing code that I missed. Bring it on.