We're coming out of stealth.
We've built our first racks after a successful A0 tapeout, $1B+ in customer contracts, and $800m raised.
Early customer tests show us achieving SOTA throughput, latency, and power efficiency on inference workloads.
Our first racks ship this summer.
good LW post on productivity techniques (link below)
went through it and checked what I actually do vs. what didn't stick
works for me
- task capture: I log every ~15-30min of work hours in my calendar and have AI generate an end-of-day summary
- underschedule: a deliberate weekly block of unplanned time (tho I got here through hypercontrol work, not productivity literature)
- focus on the wildly important: "only hell yes" rule (will write a separate post on my prioritization framework at some point...)
- rest / solitude / productive meditation: honestly my favorite category haha (jacuzzi, weighted blanket, sensory walks, breathwork, jazz clubs, baths, etc etc...)
- also: CBT and EMDR, coaching calls, strength + mobility + endurance 3x+/week, and working with a human personal assistant (not AI!)
didn't stick
- app blockers (SelfControl, Freedom)
- spaced repetition software (tried but abandoned)
the thing the post almost gets to but doesn't quite say: the post ends with "don't make yourself crazy" and flags that rigid scheduling fails where flexible planning survives
what I learned through hypercontrol work is that the urge to systematize everything can be anxiety in disguise rather than genuine productivity
so I spent quite a few years identifying which control patterns were net positive and keeping only those...
https://t.co/nzug5o2qxj
Devin Fusion is a sign of the times.
As an FDE, my conversations with enterprise CTOs and CDOs over the last two years have largely shifted from AI pilot excitement to token spend management and optimal model selection.
Fusion delivers on both with 2 new harness inventions (Sidekick, Mid-session Routing) to provide Fable-level intelligence at 35% lower cost.
Conventional model routing sucks. It passes benchmarks but fails to write code you'd actually merge.
Introducing Devin Fusion, a new hybrid-model harness for agentic coding.
In testing, it reduces the cost of Fable-level intelligence by 35% and still feels good to use.
AI should earn its keep. Introducing the AI Productivity Guarantee.
If Devin delivers less engineering value than you’re paying for, Cognition will fund your usage until it does, up to $10 million.
It’s time for the AI industry to stop maximizing tokens and start maximizing productive output.