Qwasar-Silicon-Valley-Blog-02

The Learning Companion: A New Way to Support Learning at Qwasar

Mar 19, 2026 6:59:59 AM / by Caitlin Carlton

Learning by doing and having to build real projects is hard, but skill development has always required hard work.

At Qwasar, we intentionally design programs around project-based and competency-based learning. Students spend their time working (and sometimes struggling) through real software problems, debugging real systems, and figuring out how to move forward when things aren’t obvious. That process is where learning actually happens.

Productive struggle doesn’t mean learning alone. In real engineering teams, people ask questions, sanity-check their approach, and talk through problems with others.

One of the best ways junior engineers learn is directly from senior engineers, much like a master-to-apprentice relationship. This, however, doesn’t scale in a learning setting.

That’s why we’ve created the Learning Companion Agent - an agent designed to support learners as they grow into junior engineers and enter the technical workforce. It’s always available, never sleeps, and has an extensive general technical knowledge base.

The Learning Companion: A New Way to Support Learning at Qwasar

What Is the Learning Companion Agent?

The Learning Companion Agent is a support tool built into the Qwasar platform. It understands Qwasar’s curriculum and how the platform works, and it’s designed to act as a learning companion for students as they work through projects, just as a senior engineer would assist a junior engineer on the job.

It is intentionally designed to support learning, not give answers. The agent is aligned with Qwasar’s project-based and competency-based learning model and is meant to help students think more clearly about their work, not to do the work for them.

What’s the Point of the Agent?

Project-based learning is powerful, but it can be uncomfortable. Students often get stuck, not because they’re incapable, but because they’re unsure how to start, what concept they’re missing, or how to break a large problem into smaller steps.

The Learning Companion exists to support those moments of friction. It’s meant to help students:

  • clarify concepts they don’t fully understand
  • organize how to approach a project
  • think through bugs or blockers
  • understand technical vocabulary
  • find relevant learning resources

Sometimes just being able to rant can be helpful, or treating the agent like your rubber duck.

In other words, it helps students move forward without removing the learning process itself.

How the Learning Companion Works

The Learning Companion is available within the Qwasar platform. Students can choose when and how to use it:

  • Some may consult it at the start of a project to plan their approach
  • Some are reminded to ask for help when blocked
  • Some may use it regularly
  • Others may not use it at all

There’s no required usage pattern. The agent is designed to be optional and student-driven.

When students ask for help, the agent is prompted to facilitate learning rather than provide direct solutions. Instead of handing over answers, it can:

  • ask clarifying questions
  • suggest ways to break a problem down
  • help students reflect on what they’ve tried
  • point to relevant resources
  • help clarify confusing terms or ideas

This mirrors how learning often happens in real engineering environments: people talk through problems, get guidance, and still do the work themselves.

The Learning Companion: A New Way to Support Learning at Qwasar

What the Learning Companion Does Not Do

The Learning Companion is not a coding assistant or an automation tool. It does not:

  • complete project code
  • run code
  • write full solutions
  • finish assignments
  • act as a general-purpose AI assistant

It is also limited to the learning context on the Qwasar platform. It’s not designed for unrelated questions or general tasks outside of learning on Qwasar.

Why This Matters

There’s no shortage of tools that promise to “solve” problems for learners. But solving problems for someone isn’t the same as helping them learn to solve problems themselves.

The Learning Companion is an experiment in supporting adult learners in project-based environments without removing the struggle that makes learning stick. It’s designed to encourage better learning behaviors: asking clearer questions, breaking problems into steps, seeking resources, and reflecting on what you don’t yet understand.

The Learning Companion: A New Way to Support Learning at Qwasar

An Evolving Tool

This is an early feature. Like any new learning tool, it will evolve based on how learners actually use it. The goal is to keep improving how we support project-based learning without turning learning into something passive.

Learning to build real systems is hard. The Learning Companion isn’t here to make it easy;  it’s here to make it learnable.

Tags: AI learning assistant for coding, productive struggle learning, learning companion agent, project-based learning platform, competency-based learning, junior software engineer training, coding education tool, learn to code with AI support

Caitlin Carlton

Written by Caitlin Carlton

Subscribe to Email Updates

Lists by Topic

see all

Posts by Topic

See all

Recent Posts