The good old days of data science: 5 chained models and wild variance
remilouf · x · 2026-09-18
A nostalgic jab at old-school data science: inheriting a pipeline that chained 5 models back to back, with nobody able to explain why the variance was wild — a neat illustration of error compounding in stacked-model setups.
Related event: Dev nostalgia: five-model pipeline with exploding variance(2 posts)→
More from Fun
- Anthropic engineers reportedly 'worshipping' Claude as God, eigenrobot pens parody song — eigenrobot · 2026-09-19
- Mathematician's talk abstract scooped: student used AI to derive and post proofs on arXiv first — fortnow · 2026-09-19
- GPT-6 Astra builds a flamethrower demo in Three.js from a screen recording — techartist_ · 2026-09-19
- Polymarket puts 23% odds on an orbital AI data center launching by end of 2027 — Polymarket · 2026-09-19
- "The entire tech industry is building the same thing and just changing the landing page" — prasenx · 2026-09-19
- Gamer jokes WoW's endgame is razing capitals to make room for a giant datacenter — djcows · 2026-09-19