Post by Avantika Penumarty

Senior Data Engineer (Former @Meta) | Scaled Data Infrastructure for 1B+ Users | Empowering 20k+ Engineers to think in Systems, not Tools | AI & Data Tech Creator | Open to Senior IC Roles

For 2 years I called myself a data engineer. But honestly? I was just a really good analyst who learned to schedule SQL. The moment I figured that out was embarrassing. A senior engineer looked at something I built and asked: "What happens when this breaks at 3AM and nobody sees it?" I said the alert will fire. He said okay, then what? I had nothing. Never thought about it. Not once. I had built the whole thing assuming it would work. Which is fine for a dashboard. Not fine for a pipeline that 6 teams are waiting on by 8AM. That question changed how I build everything. Analysts ask: does the data look right? Data engineers ask: what happens when it doesn't show up? That's the whole difference honestly. Not Spark. Not Kafka. Not any tool. Just learning to think about what breaks, not just what works. Took me way too long to get there. Were you an analyst before moving into DE? What made it click for you?