Olly Styles

ML engineer @ Multiverse | ML PhD from @WarwickDCS

[Olly Styles @OllyCStyles]

We’ve released a new LLM agent benchmarking dataset. Check it out here:

[Patricio Cerda Mardini @paxcema]

Introducing WorkBench: a Benchmark Dataset for Agents in a Realistic Workplace Setting

abs: https://t.co/7aOUxp2sEZ

LLM-based "agents" can interact with external systems by correctly invoking their interfaces, greatly improving over vanilla LLMs.

1/n 🧵

[Techstars Montréal AI @TechstarsMTL]

We’ve had the best time working with the guys at Atlas AI! Come check out how their AI is keeping people motivated in their workouts!

Demo Day is Wednesday: https://t.co/K0gCXDXE9x

[Olly Styles @OllyCStyles]

I had some fun with this presentation at #WPCCS. Thanks to all the organisers and other speakers for putting on such a great virtual conference. @WarwickDCS

https://t.co/fjy5b9jecV

[Olly Styles @OllyCStyles]

@tanayaguha Thank you, @tanayaguha! A recording of the talk is now available on my YouTube channel: https://t.co/fjy5b9jecV

[Olly Styles @OllyCStyles]

@MahshidMJD @WarwickDCS Thank you for organising such a fun conference, @MahshidMJD! A recording of the talk is now on my YouTube channel: https://t.co/fjy5b9jecV