My experience in the Playlab PLC

I came across Playlab as part of my exploration of educational technology as I build the next chapter of my career. They give teachers tools to build custom chatbot apps on top of all the main LLM platforms, and a community in which to share and remix them. In the universe of AI tech for education, I think their approach is unusual and thoughtful.

The interface provides a prompt construction area with guidance and templating, settings for persistent memory and documents for the LLM context window, and some other helpful tweaks to improve the standard experience of interacting with AI chatbots. (Imagine the Projects feature of Claude or ChatGPT, but thoughtfully designed specifically for education.) The community is pretty broad and has lots of interesting applications, from teaching tools like lesson planning, to student-focused ones like writing feedback.

The PLC

In partnership with Relay Graduate School of Education, Playlab offers a three-day professional development course. It introduces educators to LLM prompt construction and guides them through the Playlab interface. It's almost exclusively practice-based, with the bulk of the time devoted to designing and creating their own tool.

I decided to join the PLC for a few reasons. First, I wanted to have a direct experience to inform my thoughts and feelings about using AI in an educational setting. I also wanted to get a sense for how other educators were thinking about and using the technology. Finally, I'm exploring employment opportunities there, and getting to know their product is a critical first step.

The sessions

The PLC takes place over three sessions on Zoom, each 1 ½ hours long. It's a pretty standard format with initial setup and large group instruction and short discussion, followed by breakout sessions for focused work, and then concluded with a final whole-group reflection period.

The facilitators did a great job of running each session, and everyone was very engaged and supportive. There were lots of very creative ideas. I liked that they emphasized inclusivity and respectfulness, set good ground rules, and differentiated instruction to accommodate both novices and people with prior experience using LLMs. There were multiple opportunities to attend open office hours for extra help.

The final session was primarily devoted to showcases, in which each breakout room was a session for an educator to show off the tool they were working on. Given the short timeframe (three sessions over four weeks), it was helpful that the expectations were to demonstrate ongoing work, not a polished product.

My app

Since I have a background in coding education, I decided to focus my work there. I had several ideas coming into the first day, but none of them felt right. So I decided to turn to Claude for brainstorming. Here's my initial prompt.

I'm brainstorming AI assistants to build to make teachers' lives easier. I'm focusing on coding teachers who are addressing novice learners.

I think ideation is one of the strong points of LLMs. Stochastic language generation is one way to think about brainstorming, and hallucinations are harmless in that context. If you treat the chat as a real dialogue, you can generate and iterate on ideas in much the same way you would with a decent human partner.

After maybe 20-30 minutes of work, I had a solid idea: an aide for teachers who are struggling with how to integrate LLMs and "vibe coding" into teaching novice-level programming. This is a topic I'm wrestling with myself. I decided to narrow the focus to creative coding using Processing.js.

The two big products from my Claude session:

Check out the current version and see what you think! There's plenty of room for improvement, but I think it's a solid example of the utility of LLMs—again, squarely in the domain of ideation.

Concluding thoughts

I still have misgivings about LLMs, especially in mission-critical settings. Few things are more sacred than teaching and learning, and the relationship between student and teacher, and any technology that has the potential to undermine them should be adopted with caution and care.

Playlab is a creative application of AI to education, and I enjoyed using their tool. It's a valuable resource for both educators and students, and it's designed for creativity and discovery. I can see the influence of Scratch on the community feature, and that's a fantastic benefit on top of the tool itself. With the right precautions on the part of both Playlab and the people who use it, it has the potential to be a positive force in the inevitable transformation of education that accompanies the AI revolution.