May 2017 – Jan 2021 · 3 yrs 8 mos · Gurgaon, India

• Led engineering at India’s largest wedding platform (1M+ users).
• Scaled platform traffic from 100K → 2M+ visitors/month, app downloads 50K → 1.1M.
• Modernized stack (PHP → Node/Java, Angular → React, iOS/Android upgrades) → doubled team productivity.
• Drove 20x traffic growth, reduced costs by ₹30L/yr, improved SEO & user experience.
• Launched features like live video e-commerce, personalized widgets, boosting vendor views +50% and leads +25%.

Sep 2015 – Mar 2017 · 1 yr 6 mos · Gurgaon, India

• Introduced personalized widgets built using collaborative filtering and Marketing Automation platform to personalize user communication leading to an overall 15% increase in daily orders and an increase of 11% to 35% in the percentage of orders via personalization.
• Creation of User Persona (gender, preferred categories, price ranges, and brands) to solve the cold start problem and personalise the user experience for first-time users resulting in a 12% increase in conversions.
• Led a cross-functional initiative integrating personalization and user persona service for Search & Ads teams. This resulted in a 25% increase in first-fold order conversion for Search and a 100%+ improvement in CTR for Ads. The success prompted other teams to adopt our approach.

May 2014 – Sep 2015 · 1 yr 4 mos · Gurgaon, India

• Worked on developing cross-device tracking and targeted advertising products, including TV ads to mobile retargeting. Designed and implemented the reporting and analytics system for the DSP campaigns which helped in optimising ad spends across different parameters like city, platform, make, publisher, etc. Also developed User Profiling system to generate a segment of users for audience targeting and fraud detection.

Feb 2012 – May 2014 · 2 yrs 3 mos · Bengaluru Area, India

• Worked on following products:
• 1. MIIP Project: Serve targeted ads of products of E-commerce partners and Search Syndication related keywords as mobile ads. Revenue peaked to $100k/day with Search Syndication partners: Yahoo and Microsoft.
• Implemented K-means clustering to generate clusters of categories related keywords from the product title and descriptions involving TF-IDF, SVD for reducing dimensionality, NLP’s part of speech tagging to find relevant keywords in the process.
• Designed and developed the whole reporting and analytics system of the product. Optimized hourly jobs (were processing 50GB data/hr and ~1TB data/day) to run within 5 mins from earlier ~50 mins.
• Developed CTR prediction model using Variance Minimisation, a variant of decision tree, to choose relevant e-commerce categories and products for the user serving ~20M ad impressions/day.
• 2. AppSense: Recommend relevant ads of new undiscovered apps to the user
• Developed an algorithm to generate user-based and item-based recommendations based on the historical usage of apps in the first version and on their historical CTR in the next version leading to a 6% increase in newly launched apps. Also implemented the reporting and analytics system using Pig and Hadoop.

Sep 2007 – Feb 2012 · 4 yrs 5 mos · Bangalore

• Worked on mainly two products:
• 1. RecoEngine: an intelligent content serving engine that gives statistical recommendations using Mahout and Lucene to mobile users for subscribing to “Hello Tunes” leading to a 22% increase in monthly subscriptions.
• 2. Adwiser: Serve targeted advertisements to mobile subscribers through various mobile channels like RBT, SMS, ODP, etc bringing in ~$1.5M/yr in incremental revenue.

Aug 2006 – Sep 2007 · 1 yr 1 mo

• Worked on Manufacturing Backplane, a mission-critical application, generating a revenue of $1.4M/hr, used for the HP server in the Asia Pacific
• region. My role as SE was to handle new business requirements in Java and MySQL.