This as-told-to essay is based on a conversation with Nandita Giri, a 32-year-old software engineer at Microsoft in Redmond, Washington. It has been edited for length and clarity.
My journey began with problem-solving and math long before I knew I’d end up in Big Tech.
I studied at the National Institute of Technology, Kurukshetra, in India. Amazon, Microsoft, and Google often scout talent from this school.
Amazon hired me straight out of college, and I moved to Seattle in 2018. To get the job, there was a logical analysis test, and only a handful of us passed. Then came multiple interviews based on problem-solving. That’s how my journey in tech began.
I’m now a senior software engineer at Microsoft, and my AI skills are in high demand.
I’ve always enjoyed problem-solving
When applying for roles at Amazon, it’s not about solving problems as quickly as possible, but solving them optimally. You must carefully consider the issue so that the solution can scale and be easily maintained. Amazon has a set of leadership principles that play a key role in hiring decisions, and understanding these principles is just as important as your coding skills.
On my own time, I practiced on LeetCode, a platform for coding challenges, because I liked it, not because I was preparing for interviews. Over time, I developed a strong interest in AI.
Once I arrived at Amazon, I identified tasks and patterns that could be automated and suggested AI-based solutions to management, primarily focusing on internal workflow automation and data-driven decision support systems. That experience shaped my interest in building intelligent tools for enterprise use. Leadership gave me the green light to implement them, and I successfully integrated AI into our team’s workflow.
After working at Amazon for about four years, a recruiter from Meta contacted me on LinkedIn. I never aimed to work at Meta, but my skills opened up new opportunities. I wanted to work more deeply in applied AI, and Meta offered an opportunity to focus on building intelligent systems with large-scale data and infrastructure. It was a natural next step to grow in the AI space.
I went through the interview process, secured the job, and began working at Meta in 2022.
I’m now a senior software engineer at Microsoft
I was referred internally to Microsoft based on my work at Meta. I decided to move because I wanted to work on enterprise-focused AI products, such as Copilot, which aligns more closely with my long-term interests in building impactful tools for productivity and business transformation. I’ve been at Microsoft since 2023.
Throughout my career, I’ve had multiple competitive opportunities. With each role change, my scope of responsibility, impact, and overall compensation increased.
Most of what I know about AI, I taught myself. I spent hours outside work watching YouTube tutorials, reading blogs, and practicing. I started small, creating AI agents for personal tasks, like sending outreach emails. Tasks that used to take me a day or two can now be completed in under an hour.
Seeing those results motivated me to keep learning. What started as a personal side project has now become a central part of my career.
The work culture differs across Amazon, Meta, and Microsoft
Amazon and Meta both incorporate fast-paced learning, but Meta’s codebase is more straightforward. Facebook, Instagram, and WhatsApp are all built from a single repository, allowing you to understand the system more quickly.
Amazon’s codebase is huge, which makes the first year challenging, but the learning curve is worth it. Microsoft feels different altogether. It’s more enterprise-focused, operating at a massive scale.
I see AI as a coworker, not a threat
AI excels at repetitive or static tasks, and our job is to monitor and guide it. Managing AI, I believe, is the future of software engineering.
Demand for AI roles is skyrocketing, while traditional software engineering roles have shrunk over the last five years. Many of my friends who don’t work in AI have struggled to land new offers.
I recommend dedicating just one hour a day to learning AI. Within six months, you’ll see real progress, and these skills will be critical for the next decade. For beginners, I recommend the following:
3Blue1Brown (YouTube): excellent for visualizing math concepts behind neural networks.Fast.ai: a project-based course that helps you learn by building real-world models.Andrew Ng’s Machine Learning (Coursera): a widely recommended foundational course.Towards Data Science (Medium): accessible blogs covering practical topics and real-life applications.The Batch by Andrew Ng: a weekly newsletter that curates recent developments in AI.
AI might seem intimidating at first, but once you get confident, the opportunities are endless
If I were to advise my younger self, I would tell her to focus less on perfection and more on making an impact, to take ownership early, speak up with confidence, and prioritize learning and growth over titles.
Long-term growth comes from solving meaningful problems and maintaining resilience.
