May 2021 – Nov 2022 · 1 yr 6 mos · Bengaluru, Karnataka, India

• I was on a break to look after my physical and mental health. The break and the various recovery activities during the break helped a lot.

Jan 2019 – May 2021 · 2 yrs 4 mos · Bengaluru, Karnataka, India

• Stack - GCP, Clojure, PostgreSQL, Kafka, RabbitMQ, Redis, Docker, Kubernetes
• Gojek is Southeast Asia’s leading on-demand platform and a pioneer of the multi-service ecosystem model, providing access to a wide range of services including transportation, food delivery, logistics and more.
• I led a cross functional team of around 15 members. Our core offering was responsible for finding the best driver for a given customer’s order. The team owned 8 microservices, most of them written in Clojure.
• Designed the overall integration with Google Fleet Engine, that allowed us to achieve a significant increase in completed Transport bookings in Singapore. We collaborated closely with Google engineers to optimize the performance of their offerings in Indonesia. Overall we were able to reduce our Fleet Engine billing for Indonesia by a few million USD.
• A very common Driver Partner support ticket is “Why am I not getting a booking?”. A significant portion of these tickets had to be looked into by our team, due to the support folks not having the relevant internal information. This often led to increased ticket resolution times and constant monotonous support work for the engineers. The team built a solution to allow exposing all relevant decision making data from a critical path in our core application. This allowed the support team to quickly answer a lot more of these tickets, significantly reducing the burden of support on our team.
• As an engineering manager my responsibilities didn’t include day to day coding. However I remained hands-on to an extent, by joining an existing pair for some time as they worked on a story, conducting refactoring/design/testing demonstrations, reviewing code changes and deployment documents. I enjoyed mentoring the team and growing with them. I enabled and delegated folks to lead pods of their own. We used the DARCI framework for all our design docs, technical docs etc and I would play the decision maker role.

Jun 2018 – Dec 2018 · 6 mos · Bengaluru, Karnataka, India

• Stack - AWS, Clojure, PostgreSQL, Kinesis, SQS, Elasticsearch, JS (React), NodeJS, Docker, Kubernetes, Chef
• The Client Engineering team is responsible for building the websites and mobile apps for our customers. We created a custom web framework that encodes all the best practices for making a fast, performant, search-engine friendly website. This makes the websites fast by default, and any new research and performance improvements are immediately available to all our customers. Some of our customer websites are among the fastest digitial media publication websites in India.
• I was the common technical lead for the various teams building frontend websites. I was responsible for overseeing the development practices as well as being the bridge between the various frontend teams, the core platform team, the customer experience team etc.
• Improved turn around time and visibility on customer reported tickets by introducing new processes.
• Setup a formal training program to teach new hires concepts in JS, CSS, HTML. The attendees are far more confident and productive post the program.
• Improved Lighthouse Performance Audit Report score by analyzing and reducing the frontend JS size by over 100kb.
• Setup performance budgets for various frontend metrics and tools/processes to easily measure the values.
• Introduced Component Driven Development in the frontend teams to help create a resuable React components and follow a consistent design guideline.

Feb 2018 – Jun 2018 · 4 mos · Bengaluru, Karnataka, India

• I was responsible for setting up our performance testing initiative. We had to answer questions like, how to identify the scenarios & simulations to test, what tooling to use for running the performance tests, recording the metrics, analyzing the results, identifying the bottlenecks. The goal was to perform this exercise once, setup the pipelines and workflows and disseminate the knowledge to all the platform teams. I strongly believe each team should run/operate what they build, and knowing the performance characteristics of their apps is an important aspect of operations.
• We used Gatling for writing and running the perf tests, and wrote scripts to automate most aspects of the performance testing process. For the scenario we picked, we identified bottlenecks, which when fixed should allow us to handle 2x more traffic on the same infra.

Nov 2016 – Feb 2018 · 1 yr 3 mos · Bengaluru, Karnataka, India

• Quintype is a data driven publishing platform for digital media organizations and content creators. It has a SaaS offering, a PaaS offering and Services offering.
• The Analytics & Engagement team was responsible for building new features that help engage our customer's audience, build retention and loyalty. One of the products we built is Polltype [https://www.polltype.com]. Polltype is a SaaS app for creating opinion polls, with realtime feedback to engage better with user. It can be used indepently of our Platform. We also built some Platform features that were powered by the customer audience analytics data we collected.
• I led a cross-functional team of 6 employees. All the full stack developers were either freshers or relatively new to backend development or new to Clojure/functional programming. I mentored and taught the team well. The team members have grown a lot over the past year. The velocity has increased, they are able to catch unknowns sooner, ask questions that flesh out the problem better. They are contributing well in the respective teams they are working in currently.
• My team architected and implemented the entire analytics data pipeline, with a focus on performance and reduced costs. Be it building a scalabale collector service or performant OLAP table schemas to power the Platform features like Trending Stories, Daily Newsletter Reports, Story Expiration etc
• I worked closely with the Infra Ops team to introduce more visibility into our operations, to reduce hidden costs, reduce deployment times etc. Worked with devops tools like Docker, Kubernetes, Chef, Terraform; reporting, monitoring and graphing tools like Prometheus, Cloudwatch, Grafana, Kibana; analyzed AWS Billing Reports etc

Dec 2015 – Oct 2016 · 10 mos · Bengaluru, Karnataka, India

• Quintype is a data driven publishing platform for digital media organizations and content creators. It has a SaaS offering, a PaaS offering and Services offering.
• As part of the Client Engineering team (our Services offering),
• I helped in the creation of our Rails starter app, that new customers would use as a base.
• I was responsible for taking three of our customers live on our platform. It involved bit of project management, client interaction, team building/mentoring, day to day pairing and individual contributions as needed.
• Part of the job was to fully understand our platform and suggest to the clients how best they can use of our platform. As well as understand the client’s challenges and come with Platform features that can benefit them and many more publishers.

Nov 2012 – Mar 2015 · 2 yrs 4 mos · Mumbai

• Stack - .Net, C#, Asp.Net, JS (jQuery), WinForms, MongoDB, MS SQL Server, Powershell.
• I co-founded Firmzen [http://www.firmzen.com] in 2012. Firmzen is a SaaS practice management solution for Chartered Accountants. The company was selected to be part of the Morpheus Gang portfolio [http://morpheusgang.com]. We have over 50 active paying customers spread across the country and internationally.
• My primary responsibility was engineering. Being a small 2 person startup, I was also involved with presales, proof of concepts demos, post sales installation and customer support.
• I architected and built the core products. The app was a C# ASP.Net MVC web app with rich jQuery based frontend and MongoDB as the database. Some of functionality include ­ automatically populating employee work list, sending sms/email reminders, user access privileges, creating invoices, document upload and management, attaching custom user defined information to client profile.
• I improved the frontend and backend performance. Be it optimising the initial and subsequent page loads or profiling, identifying and improving slow web requests and MongoDB database reads/writes.
• I automated the builds, deployments, Windows server provisioning, periodic file backups, cleanups etc
• I created detailed plans for hosting Firmzen on AWS. This included planning for redundant instances, failover support, MongoDB replica set, centralized logging etc as well as writing Powershell scripts to configure and monitor the various services

Apr 2009 – Mar 2013 · 3 yrs 11 mos

• I co-founded Creative Riot [http://creativeriot.com] in 2009. We were incubated at CIIE, IIM-­A's incubator and also received seed funding from Microsoft India and IIT ­Bombay's entrepreneurship cell E­Cell. Our flagship was a B2B software product named Enpower PC [http://www.enpower.in], a centralized computer power management software. It helped save our customers over a crore rupees in energy bills [http://enpower.in/download/case­studies/Enpower­Persistent­Case_Study.pdf].
• I was primarily responsible for engineering and involved in presales, proof of concepts demos, post sales installation and customer support.
• I architected and built the core products. The products consisted of Windows Forms desktop application & system service written in C#, as well as a Asp.Net web application with a jQuery powered UI. The web application had admin functionality and reporting dashboards.
• The software clients were installed on thousands of computers across the country wide offices of a corporate client. The system was designed to not consume too much network bandwidth. It could work offline and sync data when connectivity resumed.
• I automated the deployment of our server side and client side apps as well as other tasks.