I wanted to understand the patterns behind when solo or remote work succeeds versus when teams and co-location seem to matter. This started as personal research and turned into a deeper exploration of historical examples, technology companies, and specifically AI research.

The question isn't really "remote vs in-person":it's more about understanding what type of work you're doing and what the bottleneck actually is. Deep thinking seems to benefit from different conditions than coordination-heavy execution.

I compiled data on 20+ AI breakthroughs by team size, profiled researchers who've taken different paths (from pure independent work to running large organizations), and tried to identify patterns without making sweeping claims.

Many conceptual breakthroughs came from small teams. LSTM had 2 authors. Attention had 3. GPT-1 had 4. Team sizes grew for scaling and execution, not ideation.

Work characteristics seem to matter more than industry norms. Long feedback loops, objective verification, and deep focus as the bottleneck appear to favor solo/remote work across many fields.

Researchers have found different points on the spectrum. From Radford (stayed purely hands-on) to Dario (moved to full CEO). Each represents a different tradeoff.