we have defeated daylight savings at @Letta_AI hq
all MTS receive 10k lux blasted direct to the retina 1 hour prior to sunset to avoid SAD
to experience: letta.com/join-us
This is my current agent interface - not a TUI or ADE, but a live updated Slack canvas.
My agent (or "superagent" is maybe more accurate) just shows me:
- what I am blocking
- progress bars for running coding agents
GPT models are significantly worse than the open weights frontier models (Kimi, GLM) at memory generation.
OpenAI's models have a bad tendency to create "rules" as memory, which are the opposite of generalized learnings.
Opus 5.5 leads here - still nowhere near the ceiling.
