Founder @ Zoon Robotics | Cited by DeepMind | Ex-Olympian

London, England
Joined April 2009
Robot vs GB No.1 at padel. Human won. Enjoy it while it lasts. ๐ŸŽพ
1
23
๐Ÿš€ Are Human-Led Product Demos Dead? OpenAIโ€™s latest update to their Realtime API has changed how I think about product demos. Their GitHub demo showcases a ๐—ฟ๐—ฒ๐—ฎ๐—น-๐˜๐—ถ๐—บ๐—ฒ, ๐—ถ๐—ป๐—ณ๐—ถ๐—ป๐—ถ๐˜๐—ฒ๐—น๐˜† ๐—ถ๐—ป๐˜๐—ฒ๐—ฟ๐—ฎ๐—ฐ๐˜๐—ถ๐˜ƒ๐—ฒ ๐˜€๐—ฝ๐—ฎ๐—ฐ๐—ฒ that is controlled by the user's voice. No scripts, no limitations - just an infinite interaction surface where users can explore, ask, and interact freely. Why this matters: ๐Ÿ“ˆ Human-led sales demos are inherently linear and resource-intensive. With this real-time AI capability, any software product can become an ๐—ถ๐—ป๐˜๐—ฒ๐—ฟ๐—ฎ๐—ฐ๐˜๐—ถ๐˜ƒ๐—ฒ, ๐˜‚๐˜€๐—ฒ๐—ฟ-๐—ฑ๐—ฟ๐—ถ๐˜ƒ๐—ฒ๐—ป ๐—ฒ๐˜…๐—ฝ๐—ฒ๐—ฟ๐—ถ๐—ฒ๐—ป๐—ฐ๐—ฒ - scalable, personalized, and available 24/7. The implications are huge: - ๐—ฆ๐—ฎ๐—ฎ๐—ฆ ๐—ฑ๐—ฒ๐—บ๐—ผ๐˜€ ๐˜๐—ฎ๐—ถ๐—น๐—ผ๐—ฟ๐—ฒ๐—ฑ to every userโ€™s curiosity. - ๐—•๐Ÿฎ๐—• ๐˜€๐—ฎ๐—น๐—ฒ๐˜€ ๐˜๐—ฒ๐—ฎ๐—บ๐˜€ can focus on high-level strategy while the AI handles the walkthroughs. - ๐—–๐˜‚๐˜€๐˜๐—ผ๐—บ๐—ฒ๐—ฟ ๐—ฒ๐—ฑ๐˜‚๐—ฐ๐—ฎ๐˜๐—ถ๐—ผ๐—ป ๐—ฎ๐—ป๐—ฑ ๐˜€๐˜‚๐—ฝ๐—ฝ๐—ผ๐—ฟ๐˜ can leverage this as a teaching canvas. The future of product demos? ๐—ฃ๐˜‚๐˜๐˜๐—ถ๐—ป๐—ด ๐˜‚๐˜€๐—ฒ๐—ฟ๐˜€ ๐—ถ๐—ป ๐˜๐—ต๐—ฒ ๐—ฑ๐—ฟ๐—ถ๐˜ƒ๐—ฒ๐—ฟโ€™๐˜€ ๐˜€๐—ฒ๐—ฎ๐˜. What use cases can you think of for this tech? Drop your ideas in the comments - see you down there!
1
173
Robert Playter, Boston Dynamics CEO: "Writing a draft proposal might easily be done with ChatGPT now...but on the other hand you also want it to be "right", and they don't know how to make it "right" yet" He is a huge idol of mine, but he's wrong, here's whyโฌ‡๏ธ
1
3
356
He is absolutely correct about AI disrupting bid, proposal, and tender writing. However, he might not be aware that the team at @AutogenAI has solved the issue of truthfulness in language models through hybrid generative referencing.
1
1
4
357
Our technology has helped numerous large suppliers to government secure significant contracts. @BostonDynamics, we'd be more than happy to give you a demonstration of our technology in exchange for an opportunity to pet Spot for an hour! ๐Ÿ˜‚
2
217
The State of AI: 10 Predictions for 2023 by AutogenAI ๐Ÿงต
1
1
3
293
๐Ÿ—ฃ๏ธ 10) Non-native English speakers will be able to gain traction in professional capacities like marketing and legal - where native levels of reading and writing are necessary - due to the incredible translation, comprehension and writing abilities of new LLMs.
1
206
This is so much fun! A Double Deep Q-Network solves the Lunar Lander game from OpenAI Gym after only 250 episodes of experience. @OpenAI
2