When a university with thousands of students makes each learner feel like someone is paying attention, that is not magic. It is thoughtfully designed technology.

Continuous academic support is one of the strongest drivers of student retention. The challenge is that scaling it has a structural limit: faculty simply cannot support every student exactly when support is needed. This is the story of how one university decided to rethink that limitation.

Continuous academic support: the limitations of good intentions

A question asked at 11 PM the night before an exam. A working student studying during a commute. A learner who does not feel comfortable speaking in class, but would ask in a chat. These are not edge cases — they are the daily reality of higher education across Latin America. — Romina Bertorello, Marketing Manager | Bitlogic

Faculty members do what they can. They answer messages after hours, revisit previously covered concepts, and provide support through personal effort and dedication. But good intentions do not scale. And when support does not scale, the students who need help the most are often the least likely to receive it.

For students

Questions become barriers to progress. Disconnection slowly turns into disengagement and, eventually, dropout.

For faculty

Time becomes fragmented between planning, assessment, and individualized support.

For institutions

Early warning signs remain invisible until it is too late. Data exists, but it is not transformed into meaningful action.

For the system

Attrition accumulates silently.

The question Universidad Andrés Bello (UNAB) asked was not how to hire more tutors. The real question was: how can support be available whenever and wherever students need it — without depending on a human being being available in that exact moment?

The response: aprendiz

Together with Bitlogic, UNAB implemented aprendiz — an AI-powered academic assistant integrated directly into Canvas, the university’s LMS. But aprendiz is not a generic chatbot. It is configured by course and understands each class’s specific content, learning materials, assessment criteria, and academic calendar.

What makes aprendiz different is not simply answering questions. The system identifies the learner’s level of preparedness, recommends the next learning step, and adapts responses to both the student’s context and intent.

UNAB Case Study — Chile

Before implementation, UNAB faculty spent significant time responding to student questions outside class hours, attempting to manage increasing demand without the ability to deeply support every learner. The model had become unsustainable: more students, but the same institutional capacity.

With aprendiz available 24/7, working students, nontraditional learners, and students hesitant to ask questions publicly gained access to a frictionless and judgment-free learning environment. The assistant responds with the appropriate tone, depth, and guidance for each moment of the learning process.

The results became visible quickly:

• Responses aligned with course-specific materials
• A smoother learning experience
• And most importantly, more faculty time dedicated to what only educators can truly do: teach.

“The virtual assistants implemented at UNAB represent a highly valuable resource for teaching in the biological sciences, enriching student learning while also promoting inclusion by democratizing access to advanced technological tools without bias.” — Dr. Ariel Reyes, Director of the Department of Biological Sciences, Universidad Andrés Bello – Chile

When student conversations become Institutional Intelligence

Every interaction between a student and aprendiz generates something institutions previously struggled to capture at scale: granular evidence of the learning process itself.

Which concepts generate confusion?

Which students are progressing confidently?

Which topics create spikes in questions before an exam?

The aprendiz analytics module transforms these interactions into actionable intelligence for faculty and institutional leaders:

• Real-Time course readiness

Faculty gain visibility into the group’s cognitive and conceptual readiness before exams — without reviewing individual conversations manually.

• Early identification of At-Risk students

Usage patterns, frequency of questions, and error types combine to identify students who may require support before they ask for help themselves.

• Individual progress by topic and cognitive level

Institutions can track not just activity, but how understanding evolves over time — identifying mastery, gaps, and persistent challenges.

• Faculty intervention with context

Educators gain clarity on whom to contact, when to intervene, and how to personalize support — without relying on intuition alone.

This is where the paradigm shifts:

Students are not the only ones receiving personalized support. For the first time, faculty gain real visibility into the learning process of an entire cohort — without dramatically increasing workload.

What AI makes possible — and what it does not replace

What AI makes possible

1. Speed

Thousands of interactions processed in real time, allowing insights to emerge before challenges escalate.

2. Precision

Models trained on the actual content of each course — not generic datasets.

3. Availability

24/7 support, without exception. Students needing help at 2 AM can still receive meaningful guidance.

4. Visibility

Faculty gain unprecedented insight into how students are learning across the course.

5. Human Focus

Educators spend less time on repetitive support tasks and more time on high-value teaching and mentorship.

6. Scalability

The model performs equally well for 50 students or 5,000.

What AI does not peplace

AI does not replace faculty judgment.

Technology cannot determine how to interpret a student’s situation, how to approach someone struggling academically, or how to redesign an entire class based on emerging patterns. Those decisions remain deeply human.

What changes is that educators now have better information to make them.

Hyper-personalized learning is no longer a future concept. It is already happening today — in real universities, with real students, and with educators who finally have the tools to see what was once invisible.

When technology is designed with educational purpose — not as an end, but as a means — learning journeys do more than continue: they grow, evolve, and transform.

Is your institution ready to take that step?

Let’s design the future of education together.

Let’s Talk