“We didn't want another flashcard app. We wanted a system that actually understood how students learn and gave teachers the tools to guide that process in real time. That's what Cogent Labs built.”
MemPals Team
MemPals
Challenge
Traditional study tools were built for passive consumption, not active retention.
Students have never lacked content. What they have always lacked is a structured way to engage with it at the right time, in the right sequence, with the right level of difficulty. Most digital learning tools solved the delivery problem without touching the retention problem. They made content accessible. They did not make it stick.

The students MemPals set out to serve were caught in a familiar loop. They re-read their notes. They highlighted passages. They ran through flashcards in linear order and called it revision. None of it was designed around how memory actually works. The content moved through their sessions. Very little of it stayed.

Teachers faced a different but related problem. Digital learning had given students more autonomy, but it had taken away the structured guidance that made classroom learning effective. There was no reliable way to run a live session, evaluate comprehension in real time, and feed that evaluation back into the student's study plan. The tools that existed for flashcards, session management, and performance tracking were separate systems that did not speak to each other.
The result was a fragmented experience on both sides. Students studied without direction. Teachers taught without visibility. And the gap between delivering content and actually building knowledge stayed wide open.
The tools we were using weren't connected to each other and they weren't connected to how our students actually learn. We needed something that brought all of it together.
Solution
Cogent Labs built a centralized learning system where every interaction feeds the next one.
The core insight behind MemPals was that retention and teaching are not separate problems. They are two sides of the same system. Solving one without the other produces a tool that works in isolation but fails in practice. Cogent Labs built them together.
At the foundation is a spaced repetition engine that adjusts what a student sees based on how they performed the last time they saw it. Cards that a student struggles with return sooner. Cards they have mastered recede. The system does not treat all content equally because students do not retain all content equally. Every session becomes a personalized study plan built from the student's own performance history.

Around that engine, Cogent Labs built a dual-interface system designed specifically for the two people in every learning relationship. The teacher interface gives educators full control over live sessions — starting and ending them, navigating cards, rating student responses on a five-star scale, and moving through content at a pace the class can follow. The student interface is intentionally restrained. Students see the question, wait for the reveal, and receive their grade. There are no distractions and no ambiguity about what the session requires of them.
The room-based architecture ties both interfaces together. Teachers create rooms, assign decks, and share a unique access link. Each room becomes a contained learning environment with its own assigned content and its own performance record. Session data flows back into the spaced repetition engine, so a student's performance in a live class directly shapes what they study on their own time.
Flashcard creation was built to match the way educators actually work. Cards support markdown formatting and image uploads. Decks are organized into modules. Tags and filters make retrieval fast across large content libraries. The system was designed to be used by real teachers building real curricula, not by developers configuring a tool.
The moment we ran our first live session and saw the student scores feeding directly back into their study queues, we knew Cogent Labs had understood exactly what we were trying to build.
Results
MemPals became a complete learning system where studying, teaching, and improvement happen in the same place.
Before MemPals, the learning workflow was fragmented across tools that did not communicate. Flashcards lived in one place. Session management lived in another. Performance data, if it existed at all, lived nowhere useful. Students studied in isolation. Teachers taught without feedback loops.

What Cogent Labs delivered was a single system where every part of the learning process connects to every other part. A teacher runs a live session. Student performance is scored in real time. That data enters the spaced repetition engine. The student's next independent study session reflects exactly where they struggled. The loop closes. Learning compounds.
The impact is structural rather than incremental. MemPals did not make an existing workflow faster. It replaced a fragmented set of tools with a coherent system built around how retention actually works. Teachers gained visibility they had never had in a digital environment. Students gained a study experience that adapted to them rather than requiring them to adapt to it.
Spaced
Repetition Engine — adaptive learning built into every session
Dual
Interface System — purpose-built for teachers and students
Live
Real-Time Learning — teacher-led sessions with structured student interaction
Our students are retaining more and our teachers have real visibility into how sessions are going. The platform doesn't just store content — it actively works to make sure that content stays with the student.
Conclusion
The gap in digital education was never content. It was the absence of a system that connected teaching, studying, and retention into a single continuous loop.
MemPals had the vision for what learning should look like. What they were missing was the architecture to make that vision function as a product. Spaced repetition algorithms exist. Live session tools exist. Flashcard builders exist. What did not exist was a system where all three worked together — where a teacher's real-time evaluation of a student fed directly into what that student studied next.
Cogent Labs built that connection. Not as a feature, but as the organizing principle of the entire platform. Every component was designed to pass information to the next one. The result is a learning environment that gets more useful the more it is used, because every session generates data that improves the next one.
The same principle applies to any education product trying to move beyond content delivery. The tools that will define the next generation of learning are not the ones with the most content or the most features. They are the ones where teaching and studying are not two separate acts but two inputs into the same intelligent system.
Have an idea worth building with AI?
Tell us about your workflow. We'll show you what a custom AI system can do.
Book a discovery call
