Michelle Blomberg

Render

Case study · AI Tools & Strategy · Learning Design

The library of single-shot skills behind Render, and the few agents that orchestrate them.

The single-shot skills students use all semester, and take with them. Each agent has its own tab above.

Render is training wheels. Students use these agents and skills all semester to run a real job search, and along the way they are learning to build them. At the end they take them with them: Render hands each one over as a prompt file the student runs in their own AI, such as Gemini, so they keep working after the course ends. Every student’s agents are built from their own goals, resume, and saved jobs, so no two are the same. Here is everything the tool contains.

Agents and skills, the difference

A skill is a single focused job: something in, a useful result out, one step. An agent is a role or an orchestrator: it plays a persona, runs several skills and personas together, or works on its own over time. Render is a library of skills with a few agents on top, and the student takes both with them.

The agents

Each agent now has its own tab above, showing exactly what it does and what the student gets back.

The skills

The single-shot workers the student uses all semester and exports to their own AI:

How you take them with you

At the end of the course, Render packages every agent and skill as a clearly labeled prompt file, with a plain-English readme, inside the student’s export zip. They paste each one into their own Gemini, Claude, or ChatGPT and keep using it. The Render interface does not have to live on; the agents and skills do.

Try one: the training-plan skill

A working demo, using a built-in catalog. In the full tool this runs on Gemini and draws on the student’s own saved jobs. Paste a few requirements and build a plan.

Paste the requirements from the jobs you collected

One requirement or skill per line. Not sure where to start? Load the sample.