Tackling large user traffic with Ajay Sampat, Sr. Engineering Manager, Lyft
In our first episode of Breaking 404, a podcast bringing to you stories and unconventional wisdom from engineering leaders of top global organizations around the globe, we caught up with Ajay Sampat, Sr. Engineering Manager, Lyft to understand the challenges that engineering teams across domains face while tackling large user traffic. Through this episode, Ajay shares his personal experiences and hardships that developers/engineers face in their day-to-day tasks.
Arbaz: Hello everyone and welcome to the first episode of Breaking 404 by HackerEarth, a podcast for all engineering enthusiasts, professionals and leaders to learn from top influencers in the engineering and technology industry. This is your host Arbaz and today I have with me Ajay Sampat, Sr. Engineering Manager at Lyft, a ridesharing company based in San Francisco, California.
Ajay: It’s great to be here and share my journey with the global HackerEarth community.
Arbaz: So let’s get started with a little bit about yourself? How has your professional journey been?
Ajay: BS, MS & HITACHI
- I moved from Mumbai, India to the United States when I was 18.
- I graduated with bachelor’s & master’s degrees in computer science & engineering from Ohio State & Santa Clara University respectively where I had a deep interest in how computers interacted with each other at lightning speed across the globe over the internet.
- I started my career working on block storage and supercomputers at HITACHI.
- I learned a lot from the Japanese work culture about focus, dedication, and quality.
- I knew I wanted to work on a consumer-focused product and hence took a leap of faith in online and mobile games with KIXEYE.
- I learned about growth culture and tactics from KIXEYE – building out a full stack team that focused on Growth Funnel of Acquisition, Activation, Retention, Revenue, and Referrals.
- I took those learnings to the telecommunication vertical with TextNow building out the Business Intelligence and growth teams building products on user segmentation and insights, attribution, lifetime value prediction, experimentation, user engagement.
- Currently, I head the Marketing Automation team at Lyft focusing on the top part of the funnel for strategic investments across paid and owned channels to scale both drivers and riders in a two-way marketplace.
Throughout my professional journey, I have had moments of introspection and self-discovery. I have asked myself.
- What do I really enjoy? Product Management or People Management?
- Do I want to work for a small, midsize or large company?
- What culture and values do I want the company to embody?
- What skills do I want to develop?
- What personal brand do I want to create?
Arbaz: One thing that all engineers would be inquisitive to know is, what is the biggest fear that you have, being the Sr. Engineering Manager at Lyft?
Ajay: This is not specific to Lyft but my biggest fear is not being able to create a highly functional team that delivers impact on the business. There are a lot of sub-dimensions to this but the key point I would like to highlight is the ability to hire and retain top talent in the competitive bay area market.
Arbaz: The burning question that everyone would love to know from someone working in the Lyft engineering team is: how does Lyft bring up a robust and scalable platform for managing high user traffic at certain times of the day?
Ajay: This is a culmination of years of hard work and learning from hundreds of engineers at Lyft encompassing Infrastructure, Developer productivity, and platform teams. I am fortunate to work with amazingly bright people who are passionate about their craft and the problems they are solving every day. Lyft shares a lot of in-depth articles regarding our technical challenges and our approach to solving those problems in our engineering blog – eng.lyft.com. I would also like to mention that Lyft is a major contributor to the open-source community. You can find our latest and greatest advancements in networking, security, data management at lyft.github.io
Arbaz: That’s great to know. On the personal side, what is your favorite leisure-time activity that you love to do when not working?
Ajay: Spending quality time with my son – reading him stories, taking him to park with our dog, working on puzzles and experiencing nature during our camping trip.
“This is the greatest joy of my family’s life”.
Arbaz: That’s really wonderful. Back to Ajay, the professional, one thing that all tech companies globally are looking for is to minimize technical debt. So, how do you maintain a balance of technical stability (minimize technical debt) while still delivering high-quality code?
Ajay: We like to use this question in our manager interviews. I think this depends a lot on the maturity and criticality of the feature. E.g: Tier 0 core rides API should not be held to the same quality standard of a tier 2 funnel conversion feature. In the early stages of a new feature, it is important to experiment a lot in beta, with small rollouts to gather customer feedback. This might lead to some interim shortcuts and tech debt but once it’s decided that an experiment is going to be turned into a long-lasting feature it is important to scope it holistically with test coverage, edge cases, scaling, fallback plan and so on. When it comes to mid to long term planning – it is important to view all workstreams with the same lens – engineering effort vs business impact. This requires that one is accurately able to quantify the impact of working on tech debt or the addition of a new feature and help the business make the appropriate tradeoff.
Arbaz: With all the innovation and new technologies coming up, how do you see the technical landscape changing over the next few years and how will you prepare engineering for that?
Ajay: Jensen Huang, Nvidia CEO once said: “Software Is Eating the World, but AI Is Going to Eat Software”. It is getting increasingly clear that we are moving from a Mobile-first to an AI-first world. It’s all around us from the intelligent vacuum cleaners at home to the smart cars we drive. 2 main areas that intrigue me:
- The first is AI plug-ins & IDEs like Kite and pycharm which are making coding easier and more accessible. They are significantly reducing the barrier to entry to coding and now almost anyone with basic training can build web and mobile apps.
- The Second is AutoML which is democratizing Machine learning and providing ML as a service. Just a decade ago, companies needed to build deep research science expertise to learn, develop and test various models and it took months to pick a winner. With advancements in ML libraries like sklearn, tensorflow, xgboost, and introduction of tools like DataRobot and H20.ai – not all but major resource-intensive activities of feature engineering, model selection, model training, hyperparameter tuning are getting automated, leading to faster and higher accuracy models.
These technologies will continue to make great strides in the years to come.
Arbaz: Now, taking you a few years back and trying to get the fresh graduate developer out of you here. From a candidate’s point of view, what do you think is the most challenging part of any technical job assessment or interview?
Ajay: My belief is – that for most people it is Anxiety. Let’s take a coding interview, for example, obviously, you need some basic technical knowledge of data structures, algorithms, and problem-solving to do well in a coding interview which I feel most software engineers do. Where most people suffer is they let self-doubt or anxiety get the best of them. I feel if people stay calm and focused during a technical assessment, they will be able to hear the question properly, recollect their learnings, ask the interviewer the right questions and perform their best!
Arbaz: Very well said! Taking you further back in time, what was the first programming language you started to code in?
Ajay: I got my first computer which was a Pentium III in 1999, over 20 years ago. The first programming language I coded in was HTML which was self-taught so I could build a website and have my presence known on the Internet.
Arbaz: What would be your 1 tip for all Developers, Engineering Managers, VPs and Directors for being the best at what they do?
Ajay: Albert Einstein said, “Once you stop learning, you start dying”. The technology landscape is constantly evolving. This makes it very important for everyone to stay up to date with the latest trends that interest them so they can continue to sharpen their skills. That could be the latest front end coding language, cloud service or growth tactic. Luckily, this is much easier now with the plethora of knowledge consumption mediums like blogs, e-magazines, videos & podcasts.
Arbaz: Engineers and Hiring Managers are usually thought of as really serious people who are engrossed in their work and not very social. Although we see most developers plugged in with their headphones and listening to songs. What songs or music genre best describes your work ethic?
Ajay: It has to be deep house with its high momentum and tempos. And like real work and life it has buildups and drops.
Arbaz: Lastly, If not engineering, what alternate profession would you have seen yourself excel in?
Ajay: I can see myself being in stock or commodity trading which runs in the family. Our family business has been an integral part of my childhood and has had a lasting impression on me. It has taught me the value of honesty and hard work. Trading requires constant researching, building long term strategies and relationships which I enjoy a lot.
Arbaz: It was a pleasure having you as a part of today’s episode. It was really informative and insightful to hear from you.
Ajay: Thank you for having me Arbaz and HackerEarth.
Arbaz: This brings us to the end of today’s episode. Stay tuned for more such enlightening episodes. This is Arbaz, your host signing off until next time.
About Ajay Sampat:
Ajay Sampat is a seasoned growth engineering professional with expertise in scaling companies with state of the art growth technology stacks. Ajay currently heads the Marketing Automation team at Lyft. Prior to Lyft, he started the SF office for Canadian startup TextNow and led its Business Intelligence & Growth teams, making it a top 30 Android app and top 100 iOS App, tripling their DAU and revenue. Before TextNow, he spent three years at KIXEYE building out the Growth engineering organization managing multiple successful desktop and mobile game launches. Ajay started his career at Hitachi working on block storage and supercomputers. Ajay has a BS in Computer Science from The Ohio State University and an MS in Computer Engineering from Santa Clara University.
Get advanced recruiting insights delivered every month
Get advanced recruiting insights delivered every month
Get insightful articles from the world of tech recruiting straight to your inbox
This year, as the industry went through a hiring freeze, we at HackerEarth took the time to elevate our product lineup so that…
In today’s dynamic business landscape, organizations are constantly seeking ways to optimize their talent acquisition strategies to attract and retain top performers. The…
System design interviews are becoming increasingly popular, and important, as the digital systems we work with become more complex. The term ‘system’ here…
In the dynamic world of technical recruiting, where skills testing forms the cornerstone of the selection process, we have a potent tool waiting…
I recently came across this post on Reddit, and though it was posted about 3 years ago, it wonderfully encapsulates every senior engineer’s…
This article was written by Alwayne Powell, the Senior Digital Marketing Manager at 8×8 contact centre and communication platform. You can find them…