But according to Akaash Vishal Hazarika, a Seattle-based senior software engineer with a resume that reads like a “Who’s Who” of Big Tech - Google, Amazon, and Salesforce - those days are officially over.
In a recent reflection on his eight-year career, Hazarika highlights a jarring reality: the baseline for "competence" has shifted. It’s no longer just about what you can build; it’s about how you leverage Artificial Intelligence to build it faster, smarter, and more reliably.
The "Baseline" Trap
Hazarika notes that while the fundamentals - data structures, algorithms, and system design - remain essential, they are now treated as the "bare minimum." In the AI era, these are no longer the skills that win you the job; they are simply the requirements to get an interview.The real differentiator? Hybrid Engineering.
"AI is now widely used for coding, review, and design," Hazarika points out. Because tools like GitHub Copilot and ChatGPT can handle boilerplate code in seconds, companies (especially agile startups) have raised their expectations. They aren't looking for a human dictionary of syntax; they are looking for an architect who can oversee an AI-driven workflow.
A Costly Mistake
Perhaps the most telling moment of Hazarika’s recent career wasn't a promotion, but a failure. During an interview with a Silicon Valley startup in 2024, he was given explicit permission to use AI to debug a massive codebase. He chose to go it alone, relying on traditional methods.He failed.
The "eye-opener," as he calls it, was that the interview wasn't just testing his logic; it was testing his efficiency. In some modern technical rounds, candidates are asked to ship a feature within a single hour - a task Hazarika describes as "nearly impossible" without AI assistance. The test isn't "Can you code?" but rather "Can you use every tool at your disposal to deliver results?"
The New Toolkit: What Matters Now For Fresh Graduates
If you’re looking to break into the industry or level up, Hazarika suggests focusing on these four pillars:- Engineering Judgment: AI is notorious for "hallucinating" or making logic errors. Companies now prioritize candidates who can spot where an AI-generated solution fails and where a traditional approach is more stable.
- Prompt Engineering & Debugging: Knowing how to "talk" to the model to get clean, usable code is now a core technical skill.
- The Production Mindset: For fresh graduates, it's no longer enough to have a few Python scripts on GitHub. Hazarika recommends building AI-integrated projects that are actually deployed on the cloud. Showing you understand the "lifecycle" of a model is more valuable than a perfect GPA.
- AI System Design: When designing systems today, you have to account for AI-specific trade-offs - cost, latency, and the reliability of the models themselves.
The "Hybrid" Future
Hazarika’s parting advice is a wake-up call for anyone feeling comfortable in their current role. He urges engineers to move away from the binary of being "just a coder" or "just a prompt engineer.""Be the bridge," he says.
The goal is to pair deep domain expertise—the kind that only comes from years of solving human problems - with the speed of modern AI. In this new hiring landscape, the winners won't be the ones replaced by AI, but the ones who use AI to make themselves irreplaceable.