Author
@Yuchenj_UW Yuchenj_UW
grade Bx.com/Yuchenj_UW ↗
AI-infra founder mapping frontier-lab compute dynamics to public-equity proxies
Comments on and analyzes the frontier AI landscape — model r
trader score
-0.51
hit rate
43%
mean α
-0.71%
signals 14d
5
Grade = how their written analysis reads (A best). Trader score = how their last-20 timestamped calls performed vs SPY.
Their picks, scored
Across their last 7 scored bets: 43% hit rate, -0.71% mean alpha, trader score -0.51. Their last-14d mentions, direction-adjusted, have moved -3.8% since posting (mean over 1 mentions with price data).
Recent signals5receipts included
| date (PT) | ticker | author | sent | what they said | since then | receipt |
|---|---|---|---|---|---|---|
| 2026-08-27 | · | @Yuchenj_UW | · | Reports GLM-5.3-Flash at 270 tokens/s, 10% higher benchmark quality, and one-tenth the cost. | · | tweet ↗ |
| 2026-08-26 | · | @Yuchenj_UW | · | Announces GLM-5.3 open weights in 22 hours with enthusiasm. | · | tweet ↗ |
| 2026-08-26 | NVDA | @Yuchenj_UW | -0.25 | Says GLM-5.3-Flash achieved massive usage on Chinese chips, supporting an independent AI stack. | +3.8% | tweet ↗ |
| 2026-08-26 | · | @Yuchenj_UW | · | Reports GLM-5.3-Flash beating GLM-5.2 across benchmarks at less than half the size. | · | tweet ↗ |
| 2026-08-18 | · | @Yuchenj_UW | · | Praises speculative decoding and imagines local trillion-parameter inference at 100 tokens per second. | · | tweet ↗ |
Grade is our human read-worthiness rating; trader score is a rolling 20-bet hit-rate/alpha composite — different things, often disagreeing. “Since then” is direction-unaware in the table; the summary line above adjusts for which way they leaned.