Spark's mission is simple:

Help every teacher work 1-on-1 with an unlimited number of students.

1-on-1 tutoring works. It always has.

But for decades, we've accepted a tradeoff: high-quality, personalized instruction doesn't scale - and scalable education can't be truly personalized.

Spark exists to break that tradeoff.

Unlimited 1-on-1 tutoring isn't a motivational slogan. It's a consequence of separating teaching intelligence from teaching time - something that's now possible with AI.

Spark leverages AI to effectively clone a teacher's judgment, context, and teaching decisions for every enrolled student.

This will allow teachers to focus on what machines fundamentally cannot:

• Empathy: understanding how a student feels, not just what they know

• Compassion: responding to fear, frustration, and self‑doubt with care

• Vision: helping students see who they can become, not just what they should learn next

• Meaning & motivation: connecting learning to purpose, values, and real life

• Trust & relationships: the human bond that makes students show up and persist

Spark doesn't replace teachers. It removes the repetitive work that prevents them from doing more of the above.

If you've used Spark so far, you might think of it as a course creator.

That's a natural first impression - because that's where the system begins.

Now let's step back and look at what Spark is actually designed to do.

This is a long post.

In a hurry? This image gives you a summary.

Table of contents

• TL; DR

• How Spark works

  • 1. Who is an SME?

  • 2. How does the SME teach Spark what they know?

  • 3. How does Spark teach the SME's students?

    • a) As a structured course

    • b) As an on-demand 24/7 tutor

    • c) As a coach

  • 4. How does Spark work 1-on-1 with each student? ( …

    • Step 1: Understand the student's background and go …

    • Step 2: Adapt Spark to the student

• Putting it all together

• Conclusion

Spark is an AI Tutor that subject matter experts can hire to teach their students 1-on-1.

Let's break that down.

1. Who is a subject matter expert (SME)?

2. How does the SME teach Spark what they know?

3. How does Spark teach the SME's students?

4. How does Spark work 1-on-1 with each student?

1. Who is an SME?

A subject matter expert is anyone with deep knowledge and skill in a domain. This includes:

• High school tutors

• University professors

• Domain experts within enterprises

• Independent coaches

2. How does the SME teach Spark what they know?

The SME doesn't need to learn a new workflow.

They simply capture themselves teaching - exactly as they already do.

This can include recordings of meetings, presentations, workshops, live classes, podcasts, or YouTube videos.

The SME shares these recordings with Spark. Spark analyzes the content to infer:

• The subject, subfields, and key ideas being taught

• The teacher's depth, style, and instructional approach

• The intended audience - including age range, background, and goals

3. How does Spark teach the SME's students?

Spark works the way a strong human tutor would - just without the time constraint.

First, it does the preparation work:

• Breaks the content into a modular lesson plan

• Generates structured study notes for each section

• Creates quizzes that assess understanding at multiple levels: section, module, individual recordings, and across courses

Once this groundwork is complete, Spark supports students in three distinct ways.

a) As a structured course

This includes:

1. A micro-learning experience with short video sections (typically a few minutes each)

2. Detailed study notes for every section

3. Slides extracted from the most important video frames

4. Embedded MCQ quizzes, distributed throughout the course as knowledge checkpoints

5. Auto-graded assignments that require students to apply what they've learned - through coding exercises or long-form writing (Coming soon ⏳)

b) As an on-demand 24/7 tutor

This is the AI Chatbot.

In this mode, Spark combines:

• The reasoning ability of a highly capable AI system

• Inferred context about the student's background and goals

• A strict constraint to stay within the teacher's actual content

Every response includes citations pointing to the exact source material used.

This gives students confidence that the explanations are grounded in the content they're meant to study - not generic answers.

c) As a coach

This is Spark's proactive mode.

Here, the AI takes partial responsibility for the student's progress by:

• Nudging the student toward the next smallest meaningful step (e.g., a 3-minute section or a 5-minute quiz)

• Sharing progress reports with students and other stakeholders (teachers, parents, managers)

• Highlighting strengths and identifying areas that need more attention

4. How does Spark work 1-on-1 with each student? (On the roadmap ✨)

Before a human tutor begins teaching a student 1-on-1, they typically do three things:

• Prepare a rough lesson plan

• Understand the student's current level and goals

• Adapt the plan based on the student's needs

Spark follows the same logic.

It personalizes the experience across all three roles it plays: course creator, tutor, and coach.

Note: The personalization features below are on our roadmap ✨

Step 1: Understand the student's background and goals

Spark gathers context about the student relative to the subject being studied.

For example, in a life coaching course, a student might share a summary of their background along with goals for the next few years.

This context allows Spark to work with the student in a genuinely 1-on-1 manner.

Step 2: Adapt Spark to the student

This is done by personalizing each aspect of Spark for each student.

Personalized course creation

• Customize the lesson plan by omitting irrelevant chapters or reordering content

• Coordinate multiple courses within the same subject into a coherent learning path

Personalized quizzes

• Frame questions in the student's specific context

• Adjust scenarios based on age, experience, and real-world constraints

Personalized assignments

• Design application-based tasks tailored to the student's real-life situations

Personalized 24/7 tutor

• ChatGPT already does this today.. giving you personalized answers to your questions, based on what it already know about you.

• But there's a critical difference in Spark: All conversations are strictly limited to the context of the courses the student has enrolled in.

Personalized coach

• Effective coaching requires empathy.

• With richer context about each student's background and goals, Spark can deliver guidance that feels supportive rather than robotic.

• This idea is still evolving, but the direction is clear - and we'll refine it through real usage and feedback.

At a high level, Spark does three things:

• It captures how great teachers think, explain, and sequence ideas.

• It removes time as the bottleneck by letting that teaching intelligence work in parallel.

• And it delivers that intelligence to each student in a way that still feels personal, contextual, and human.

That's what makes unlimited 1-on-1 tutoring possible - not as a promise, but as a system.

If you've already explored Spark, you'll know we're still early.

Unlimited 1-on-1 tutoring for the world's best teachers is a big mission.

We're removing the biggest constraint that's held great teaching back for decades: time.

Now it's about execution - shipping, learning from real classrooms, and getting better with every iteration.

So..we're just getting started.

And if this mission resonates with you - whether you're a teacher, an operator, or someone who cares deeply about learning - we'd love to have you along for the journey.