A conversation with Ravi Maguluri, CTO, Sify Technologies North America, on how contextual learning closes the gap between needing to know something and knowing it.
Come across something at work you don’t know how to do, and the same pattern tends to follow. You stop, go looking for the answer, learn it, then return to the task to apply it. Every step between not knowing and knowing is friction, and friction costs time and breaks the flow of work. What’s changing now is that the distance between not knowing and knowing is starting to disappear: the answer can arrive inside the work, the moment you need it, drawn from your organization’s own knowledge and correct for the situation in front of you. When that happens, learning and working stop being separate things. At Sify, we talk about this as contextual learning. I sat down recently with our CTO, Ravi Maguluri, to get his view.
Answers used to live outside the work
When I asked Ravi where he sees contextual learning showing up already, he gave an example from the day before. He’d been in Google Cloud, trying to set up a complicated configuration. A few years ago that would have meant working through manuals and searching documentation for the right steps. This time he described what he was trying to do, and an AI assistant walked him through where to go and what to run, without him having to leave the screen he was working in.
Help at the point of need isn’t new; performance support in apps and tools have been around for years. What’s new is what AI brings to it. Older performance support still put the work on you: find the right article, read it, and figure out how it maps to the problem in front of you. An AI assistant closes that last gap, interpreting your situation and answering it directly, in conversation. Ravi sees this turning up across industries. A bank associate working out whether an applicant qualifies for a loan. An insurance rep judging whether a customer should file a claim at all. In each case the guidance appears where the work is happening, rather than somewhere the person has to stop and go find it.
Learning in the flow of work
Performance support is only part of it. What’s interesting is the larger shift behind it: developing people and helping them in the moment is converging.
Some learning will always stay separate. Compliance, certification, and regulation still call for formal, deliberate training, and people still need to be signed off as qualified. But much of the rest is moving into the work itself. The skills someone would once have left their desk to learn now reach them as they do the job, so ‘learning’ and ‘needing to know’ start to happen at the same time.
This is what the field refers to as learning in the flow of work, and it’s the heart of contextual learning. The idea is straightforward. Rather than sending people away to learn and hoping they remember it later, you put the knowledge where the work happens and let them draw on it exactly when it counts. Development stops being an event on the calendar and becomes part of doing the job well.
It only works if the knowledge is sound
All of this depends on AI to deliver it. And for an AI tool or assistant to be useful at work, its answers have to be correct – reliably, every time. A general-purpose AI tool has a well-known weakness: ask it something and it will often give you a confident, fluent answer that reads as right but isn’t. AI tools are famous for hallucinating. That’s not a big problem if you’re drafting an email (and you recognize the flaw!). But it’s a big problem if you’re telling a customer how to fix something, or quoting them a price or a policy.
The way through is to stop the AI from answering out of the open internet and ground it in the organization’s own material instead: its policies, its workflows, the record of how past situations were handled and why. The AI assistant answers only from what the company actually knows, and nothing else. As Ravi put it, “in a work situation, you cannot make it up.”
That same approach answers the risk question every leader raises. If our knowledge and our people’s work are constantly flowing into AI, could it leak out? The exposure comes from sending your information to a system owned by someone else. Keep the knowledge, and the AI that reads it, inside your own environment, and there is nowhere for it to escape to. Ravi calls this sovereign technology: holding your own knowledge instead of parking it in a system you don’t control.
What it means for the people who build learning
Organizing a company’s knowledge this way turns out to be familiar work. Pulling together information that lives in a dozen formats and places, making sense of it, distilling it, and showing how it applies to a real situation is exactly what instructional designers have always done to build a course.
So the craft doesn’t disappear as learning moves into the flow of work. It changes shape. Instead of producing courses, instructional designers increasingly organize the company’s knowledge into the connected, structured form an AI can draw on reliably. Turning the raw sprawl of corporate information into something people can actually use matters more than ever.
What AI really solves is friction
The shift in the instructional designer’s work is one small example of something much bigger. What AI genuinely changes is the friction between having an idea and seeing it take shape. Anyone who has used these tools well knows the feeling: a thought becomes something you can look at within minutes, even if it takes a dozen tries to get it right. The value isn’t only speed. When your thinking materializes fast, you can judge it while it’s still fresh and push it further. The distance between a thought and a first version of it is almost nothing now.
Contextual learning is solving for the same kind of friction, just closer to the work. The old distance between coming up against something you don’t know and getting past it, the stopping, the searching, the leaving and coming back, is collapsing in the context of the workplace. The help comes to you, from what your organization already knows, and a person stays in charge of what to do with it.
Organizing what your company knows so it can reach your people in the moment they need it is the work we do at Sify. If you’re weighing what that could look like across your teams, let’s talk
