“A ship in harbor is safe, but that is not what ships are built for.” — John A. Shedd

From my time at early-stage startups, I always wanted to work at a company that had achieved a product-market fit yet was as nimble as a startup. Why? Because it is one thing to read about building for scale and another to actually do it amidst challenges unique to hypergrowth companies.

In the past year, we at Blinkit (formerly Grofers) have successfully pivoted to 10 min delivery, grown our order volumes by 3.5x (in the last two months alone), and doubled the size of our engineering team. Along with scaling our existing systems, this required solving unseen problems and building new systems from the ground up.

My journey at Blinkit started when I joined the Replenishment Engineering team last year.

The Team

Our team builds tech to improve product availability on the platform — from sourcing the right assortment to replenishing the stores for 10 min delivery magic; our systems decide what, when and how much to order and then place orders directly in vendors’ ERP systems using Electronic Data Interchange (EDI).

I built a lot of context on our business and technical architecture during my onboarding, and people at Blinkit went the extra mile (thanks Keyur) to help me ramp up quickly. To get my hands dirty, I paired with my teammates to build a constraint optimization capability in our stock transfers — to select the best possible assortment and the ideal quantity to transfer based on the current storage, manpower, and fleet capacity.

Learning and Impact

During the same time, a data-intensive job that gives insights on the stock-outs that happen on our platform started failing. While debugging this with my teammate Anmol, we found that certain pods were going out of memory during Pandas operations. In fact, even SQL queries were timing out.

I tuned the job using batched queries, efficient indexes and swapped SQL and Pandas for some data joins. Its runtime was cut down by 80%, and we got rid of the massive load it used to cause on our DB.

But my first org-level impact came a month later when I was given the responsibility to lead the team. In the months that followed, we went from 4 to 12 engineers with multiple pods and solved problems at warp speed. I couldn’t have done it without my first manager at Blinkit— Karthik, who coached me and helped me get there.

Of course, there were times when we made mistakes.

We worked on a new capability that took four months to build, but by the time we were prepping for rollout, the ground reality had changed. Unfortunately, it wasn’t going to add value to our new business proposition.

This experience came with an important realization — if our technology evolves slower than the supply chain, we’ll always play catch-up. It taught me that when things move fast, make quick decisions and execute even faster to get the first cut out the door.

“he enemy of a good plan is the dream of a perfect plan.” — Carl von Clausewitz

Another thing I learned from my teammates, is to be laser-focused on solving business problems and not marry yourself to a particular language or tech stack. One day someone would be working to create a beautiful UI, and the same person would be writing handlers for an Event Sourced application the next day.

One of the 0–1 problems we’re solving right now is Continuous Replenishment (CI/CD anyone?)

Similar to how deploying code once a week doesn’t work in an agile environment, we can’t serve all possible customer demand by replenishing stock at the Dark Stores only a couple of times a day. We need to ensure there is a constant flow of inventory from warehouses to the Dark Stores to always have relevant products in stock. In the past couple of months, we’ve built capabilities that allow seamless configuration of the supply chain for multiple replenishments along with on-demand triggers when availability is running low.

Culture

I’ve lost count of all the awesome things I’ve learned at Blinkit in the last year, but if you ask me, what’s the best part of working here? The people. There are some very talented folks with a high sense of ownership who only pick audacious problems.

It’s in the company’s culture that everyone tries to solve the biggest problems for our customers regardless of their title and Albi pushes us to raise the bar for ourselves.

Most of us have visited Dark Stores and warehouses, and some have even delivered orders to customers to get the firsthand context of on-ground operations.

This exposure is critical as it teaches us to be more empathetic and build people-first solutions for our customers and the supply chain workforce. Also, we follow the mantra “Be sincere, not serious” (credits: Naina) and make sure we don’t miss out on the fun.

I’ve learned a ton and had a lot of fun in this short time, but there is a long way to go. I look forward to building Blinkit into a world-class technology company in the years to come.