Let's get building!

Dear Convofriends! I'm back from my short break in New York, so that means catching up with everything that has happened online. And you know? The beauty of being away, a change of scenes in the real world, it's just such a lovely antidote against all the news and hype and fomo that's happening online. The amount of time it takes to do real travel, the waiting, the boredom, the normalcy...it made me re-realise that there's no way that we can keep up in real life with everything that's happening online. And it feels great! :-) So what has happened in the last week? Here's my quick catch-up!

Voiceflow PromptHacks

You probably already saw my videos on your timeline: I had a great time learning generative AI and prompt engineering in the Voiceflow PromptHacks hackathon. And speaking of 'normal': to have the power of generative AI in a familiar convo AI interface was really, really useful. If you haven't given it a try yet, by all means do!

ConvoLab!

OK OK, I'm really excited about this one: I'm starting up ConvoLab, our place to co-design, build and experiment with generative conversational AI! In our last Convoclub session, we all felt we'd like to continue building our PromptHacks projects, and thanks to Braden Ream from Voiceflow, we can!

ConvoLab is a series of regular co-creation sessions where we meet and build on our projects together. I noticed that many people in our community feel like a group can help them to keep themselves accountable, to really sit down and put in the work. And that's exactly the space that I'm creating for you.

If you're interested in participating, head over to www.convo.club and fill out the interest form!

In the news this week

It's too much, as usual, so here are some of my personal highlights. Langchain, for starters, or rather, Langflow, the visual interface for Langchain! If you haven't heard of Langchain, no worries. It's basically a bit of magic, really. In its original form, it's a Python library that allows you to integrate a LLM with all kinds of other components through APIs. That way, a stand alone model all of a sudden becomes a super-helpful component in a whole chain of other components that can all be operated through prompts. And that's really useful when you want to have natural language interactions over something that might be more rule-or decision based, like business logic.

Langchain is a Python library, and as such might not be very intuitive for most of us. And that's where Langflow comes in (OK, you still need to install it in Python, but that's something that's not too complex, and teachable. If I can learn it... ??). Once you've got it up and running, you'll see an interface that might look slightly familiar. With it, you can literally 'chain' agents, prompts, APIs and functions to create LLM-based flows of events and interactions. I'm definitely planning some evenings to learn & build!

From my Linkedin favorites

About Maaike Groenewege

Maaike is a conversation design lead, linguist and DesignOps coach with?Convocat conversational expertise. She helps both starting and more experienced conversational teams in optimising their conversation design practise, NLU analyses and team communication. Her main focus right now is on how LLMs can benefit enterprise conversational AI.

• Join?Convoclub, my boutique community for conversation designers, linguists and language lovers. We meet every other Monday at 18:00 CET, find the invites in the Convoclub Events section, or keep an eye out on LinkedIn!
• Find your next book on conversational AI in my?Convocat library