~/talk

AIE Miami Keynote & Talks ft. OpenCode. Google Deepmind, OpenAI, and more!

on YouTube 07h 54m
Talkat AI Engineer Europe 2026 · 2026-04-08..04-10· [ai-assisted-development]

speakers

name:OpenCode
links:none on file
name:Google DeepMind
links:none on file
name:Gabe Greenberg
links:none on file
name:Dax Raad
links:none on file
name:Dexter Horthy
links:none on file
name:Max Stoiber
website:https://mxstbr.com

resources

video:https://youtu.be/6IxSbMhT7v4
slides:none on file

summary

Conference claims a few concrete wins: a multi‑agent Orchestrator shows an 8.4% overall lift over GPT‑5.4 on SweptBench Pro after hotel‑room benchmarks; OpenRouter reports ~75 trillion tokens/month, 5M users and ~300 models; agent sessions run ~11× longer and tool calls hit ~99% success. Demos use SSH + Tailscale + tmux for remote coding agents, Veo/Nano Banana/Lyria for gen‑media, NVFP4 quantization on Blackwell for faster inference, and Cloudflare isolates + code mode / MCP for safe server‑side execution. A working free personal assistant runs as an MCP app.

AI-written summary · about

chapters (20)

  1. 0:00Introduction
  2. 3:03Opening remarks by Ethel Zhang and Iman Haji
  3. 7:08Gabe Greenberg (G2i): conference origin story and Orchestrator AI announcement
  4. 17:22Dax Raad (OpenCode): "You Don't Have Any Good Ideas", product restraint against software bloat
  5. 33:20Dexter Horthy (HumanLayer): what RPI got wrong, from giant prompts to control flow
  6. 56:39Max Stoiber (OpenAI) with Erik Thorelli (CodeRabbit), Sunil Pai, and Ben Vinegar (Modem): DX taste for agents
  7. 1:17:12Break: Morning coffee
  8. 1:58:36Ben Vinegar (Modem): remote coding agents over SSH, Tailscale, and tmux
  9. 2:23:19Shashank Goyal (OpenRouter): real usage data on cheaper inference and agent workflows
  10. 2:46:46Nnenna Ndukwe (Qodo): AI code quality gates across the SDLC
  11. 4:28:34Break: Lunch
  12. 4:36:38Geoffrey Huntley (Latent Patterns): software development below minimum wage and model first economics
  13. 4:57:52Philip Kiely (Baseten): safer fast inference with NVFP4 quantization on Blackwell
  14. 5:19:47Alisa Fortin & Guillaume Vernade (Google DeepMind): Veo, Nano Banana, and Lyria demos
  15. 5:46:50Anna Juchnicki (Pinterest): governed Snowflake Ops agents with LangGraph, MCP, and PR review
  16. 6:09:04Break: Afternoon
  17. 6:47:45Kent C. Dodds: Cody, a free personal AI assistant over MCP
  18. 7:09:56Rita Kozlov (Cloudflare): isolates and code mode for billion scale agent infrastructure
  19. 7:32:32Ben Davis (T3): choosing AI SDKs and generating software from natural language Markdown
  20. 7:50:37Closing remarks by Ethel and Iman

related talks

~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 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
?
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.