~/talk

Krishna Rangasayee, SiMa.ai | Supercloud 4

Talkat Supercloud 4 · 2023-10-24..10-25· [machine-learning, embedded-and-hardware]
links:none on file
slides:none on file

SiMa.ai claims a machine-learning system-on-chip that delivers L2+/L3 vehicle-class inference at 50 watts for L2+ versus 300–600 W peers, and asserts 10x performance per watt over legacy architectures. The product combines Arm TrustZone and Synopsys IP for boot-to-run security, an ML SoC silicon taped with TSMC, and a software stack that runs any CV model and a no-code Edgematic pipeline to build, compile and deploy real-time edge inference.

AI-written summary · about

  1. 0:00Introduction to edge AI
  2. 1:05Founding SiMa.ai
  3. 3:55Latency challenges in edge AI
  4. 4:59Hardware software co-design
  5. 7:29Edge market potential
  6. 8:53Key use cases today
  7. 9:55Autonomous vehicles reality
  8. 13:10Chip design innovation pace
  9. 16:05US chip manufacturing concerns
  10. 17:28Privacy and security at edge
  11. 20:15ML SOC architecture benefits
  12. 21:40Software centric approach
  13. 23:27Market opportunity at edge
  14. 25:54Future of edge innovation
  15. 27:18Closing remarks
~/search
type to search 6,518 conferences, 143,918 speakers and 192,133 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
← →
on a talk: 5 seconds back / forward
0–9
on a talk: jump to 0%–90% of the recording
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.