How an Oracle Sr. Engineer Stays Ahead of AI

Today we're doing something a little different: we're having a conversation with Nishal Pattan, a Senior Software Engineer at Oracle.

For his day job, he builds AI agents for enterprise knowledge search. (He has also been learning with Educative since his college days).

Instead of chasing every new release, Nishal shares how he decides what's actually worth learning versus what's just noise. The focus is on staying relevant in the AI era: leaning on fundamentals like System Design and architecture, applying a zero-trust mindset to what AI builds, and learning hands-on rather than drowning in theory.

If you've been feeling the AI pressure, but want a working engineer's honest take on how to keep up without burning out, you're going to get a lot out of this one.

As AI reshapes what it means to be a software engineer, staying current takes more than scanning headlines. It takes deliberate, hands-on learning.

Few people embody that better than Nishal Pattan, a Senior Software Engineer in Oracle's OCI Enterprise Engineering org, where he builds AI agents to power enterprise-wide knowledge search and help employees troubleshoot day-to-day IT issues.

We caught up with him to hear how he's navigating the AI era and where he's heading next.

Learning By Doing, Not By Theory

When Nishal found out he'd be working more on AI agents, he didn't reach for dense documentation. He reached for the kind of learning that mirrors real work.

"I relied on Educative courses, including Mastering OpenAI Codex for Agentic Coding and MCP Fundamentals for Building AI Agents which really helped me nail down the basics of AI agents, MCP servers, building skills, and plugins," he says. "These courses are more hands-on, which is what I want, I always prefer learning by doing rather than all the boring theory."

That preference for application over abstraction is a thread that runs through everything he does. The courses he values most, like a recent one on Agentic AI, are the ones whose exercises "simulate real day-to-day work."

Beating Decision Paralysis

Ask any engineer about keeping up with AI and you'll hear some version of the same problem: there's simply too much of it.

"AI-related tech is being updated at an exponential pace these days," Nishal admits. "Sometimes I face decision paralysis about what to learn."

His answer is to lean on signal over noise. He looks at the courses other engineers are using to upskill, and reads Educative newsletters to stay current. "They really help me make decisions and choose the skills I need to learn." Left to a blank Google search, he says, he'd likely end up paralyzed, "having everything organized for me adds real value in learning the skills that matter most."

What The AI Era Actually Changed

For Nishal, AI is no longer a nice-to-have productivity boost — it's the default in his daily work. But he's clear that this raises the bar rather than lowering it.

"The emphasis is no longer just on writing clean code," he explains. "It's more on architecting systems, developing with clarity about what's needed and what's not, applying a zero-trust principle to how AI builds things, and doing deep reviews. AI has made us think more deeply."

If anything, fundamentals matter more now than ever. "Anyone can leverage AI agents, but if they don't have strong skills in distributed systems, architecture, and debugging, it impacts the quality of the software they build with AI. You should be able to architect and deeply understand the software you're going to build, at a much more granular level."

It's a balance he frames as a craft in itself: "I feel it gives us the ability to solve more problems, but the trade-off between quality and velocity is something a good software engineer can navigate."

So What Is An AI Agent?

For readers who keep hearing the term but aren't sure what it means, Nishal has a simple analogy — and a caveat.

"In the beginning, it was more like a pair programmer for you. Now it has gotten so good that you can delegate work to it. In simple terms: you come up with a cooking recipe and give it to an agent who is really good at following instructions and cooks the dish for you. But the catch is that if you're not sure about the recipe, the agent can help you brainstorm and put one together." 

From summarizing your daily emails to shipping production-ready code, he sees agents being useful across the board.

Where He's Headed Next

The skills that grew his career, Nishal says, were "System Design, problem solving, digging deeper into the tech I'm working with, and a curiosity to keep learning." That same curiosity is pointing him forward: he's now planning to go deeper on RAG-based systems and multi-agent architectures.

He's blunt about why he keeps choosing Educative over the alternatives he's tried:

"I can customize exactly what I learn, and it's always hands-on — that's why I keep coming back. It's affordable, and I can learn at my own pace." The AI mock interview and System Design content even helped during his most recent job search.

The lesson underneath it all is a quietly demanding one. In an era where anyone can prompt an agent, the engineers who pull ahead are the ones who still understand – deeply – what they're building and why.

Stay curious. Stay relevant.

Nishal Pattan Senior Software Engineer @Oracle || AI Agents, MCP, RAG || OCI Enterprise Engineering