Apr 2022 – Aug 2024 · 2 yrs 4 mos · Bengaluru, Karnataka, India

• 1.Implemented a dedicated pipeline tailored specifically to cater to the needs of our Lower Affluent clientele, notably the Shopsy Customer base. This innovative approach entailed meticulous segregation of data and context at various pivotal stages spanning across four distinct layers. By adopting this methodical separation, we ensured that the recommendations generated were not just precise but also finely attuned to the unique preferences and requirements of this demographic
• Technology used : MapReduce, Spark
• 2.Pioneered the implementation of multiple recommendation widgets on the Shopsy homepage, introducing personalized product suggestions based on users' browsing history and prominently featuring bestsellers and top-rated items from their explored categories, resulting in heightened user engagement and increased sales conversion rates.
• 3.Created a new Recommendation Widget for Product Page targeting hyper-value customers. Displays only lower-priced product recommendations compared to the viewed item. Involved data aggregation and precise recommendation selection for optimal user experience.
• 4.Developed a customized content provider for the Infinite product feed from inception to deployment in just 2 months, navigating contracts with 5 upstream and downstream partners for seamless integration. Led multiple rounds of non-functional testing to optimize system scalability, supporting up to 3x BAU traffic. Achieved enhanced visibility with placement on Flipkart and Shopsy homepages.
• Technology used : Java 17, SpringBoot, Aerospike, Data Structures, Analysis of GC logs, JFR dumps.
• 5.Transforming the product feed content provider into a config-based system involved identifying parameters from various sources and making them dynamic and configurable for different page types and experiments. The aim was to enable plug-and-play functionality without the need for additional coding.

Jan 2022 – Apr 2022 · 3 mos

• Admin Tool for Centralized Workflow Management System for Advisors: Component Owner
• of the Admin Utility built as a Microservice for the Business team to create automatic
• request types. Ability to make Dynamic Forms for each request. Reduces the turnaround
• time for business as dependency on the Tech team is not present.
• Technology used: Java Spring boot, React.
• Alerts & Notification Dashboard for Advisors: Presents as one stop to advisors, helps to
• know the exact item to be worked upon and exact status of events taking place for the
• accounts in view. Built as robust system to have the ability to receive input from at least 25
• different upstream via Kafka, does various enrichments before making the view available
• to advisor. The model supports at least 300 different types of events.
• Technology used: Java Spring boot, Apache Kafka, Elastic search.

Jul 2019 – Jan 2022 · 2 yrs 6 mos

May 2018 – Jul 2018 · 2 mos · Mumbai, Maharashtra, India

• Problem Statement: Automation of the Comparison of consolidated bank
• Statements. Comparing docs in PDF format & extracting text using the OCR (Optical
• Character Recognition) Technology: TESSERACT. Conducted Document layout
• analysis using OpenCV for intelligent extraction of text & achieved about 90%
• accuracy. Used Levenshtein Distance Algorithm and Diff Algorithm for giving best
• results for comparison.

May 2018 – Apr 2019 · 11 mos · Surat, Gujarat, India

Jan 2018 – May 2018 · 4 mos · Surat, Gujarat, India

Jun 2017 – Jul 2017 · 1 mo · Ahmedabad, Gujarat, India

Jul 2015 – Jun 2019 · 3 yrs 11 mos · Surat Area, India