Jun 2023 – Oct 2024 · 1 yr 4 mos · London, England, United Kingdom · On-site

• Personalising Buyer Journey on FB IG Shops

Jun 2022 – Sep 2024 · 2 yrs 3 mos · London, England, United Kingdom · On-site

• Recommendation System For Commerce
• ➢ The ARTS ranking platform is a powerful tool for commerce businesses that seek to provide personalized and relevant shopping experiences to their customers.
• ➢ Developed the ARTS (amplified reason to shop) ranking platform to help the commerce relevance team provide personalized and relevant shopping experiences to users. My leadership in cross-functional teams led to rapid prototyping, large-scale experimentation, and fast iteration while ensuring good code quality, maintainability, and efficiency.
• ➢ Leveraging machine learning algorithms, ARTS is capable of understanding the most effective placement of shopping labels, such as "Sale" or "Free Shipping", on a product banner to highlight great shopping deals to users. This results in higher engagement, conversions, and customer satisfaction.
• ➢ Provided easy-to-use config-driven A/B experimentation support and key business metric calculation to measure the success of the system.

Nov 2021 – Jun 2022 · 7 mos · London, England, United Kingdom · On-site

• Creator Affiliate Recommendation System
• ➢ The Creators Affiliate Recommendation System is a powerful tool for affiliate creators seeking to find and promote the best products and brands to their audiences.
• ➢ Developed and implemented the Creators Affiliate Relevance workstream, which leverages cutting-edge machine learning techniques to help affiliate creators discover great products and brands..
• ➢ My leadership of cross-functional teams was instrumental in surfacing the recommendation on the Instagram app (affiliate creator dashboard), improving the visibility and usability of the system.
• ➢ To enhance the system's capabilities, I also developed and executed several key product pivots, including Suggested Products, Trending Products, and Followers Might Like Products. These pivots were designed to optimize the relevance and personalization of product recommendations, resulting in higher engagement and conversion rates.
• ➢ As a better engineering champion for the team, I ensured that the system was built with the highest standards of engineering excellence, including maintainability, scalability, and robustness.

Jan 2019 – Nov 2021 · 2 yrs 10 mos · Singapore

• Food Demand Forecasting and Simulation Platform:
• ➢ Food Demand Forecasting and Simulation Platform have enabled businesses to make more informed decisions based on accurate demand forecasting, leading to optimized inventory management, reduced food waste, and increased profitability.
• ➢ Designed and developed a cutting-edge forecasting service for predicting food signals and detecting anomalies, which is a critical component of the Food Demand Forecasting and Simulation Platform.
• ➢ Developed a generic Flink operator that consumes signals from Kafka topics and ingests them into a time-series database, which enabled the platform to handle large-scale data processing and storage requirements.
• ➢ Implemented Scheduled Forecasting (performed every x minutes) and Forecast by Request functionalities, making it easy for users to access and benefit from the platform's forecasting capabilities.
• ➢ Developed a user-friendly interface that enables users to publish demand signals to a Kafka topic and choose from a bank of pre-trained models to enable forecasting for the required signal. This approach not only simplifies the forecasting process but also enhances the platform's usability and accessibility for non-technical users.
• Food Promotion Recommendation System:
• ➢ Food Promo Recommendation System provides grabfood merchant partners with a powerful and effective tool for driving sales and revenue growth.
• ➢ Formulated a two-step approach that involves prediction and assignment, which has resulted in improved accuracy and better recommendation outcomes.
• ➢ Detailed exploratory data analysis, prepared and processed the training data using PySpark and Scikit-learn, and built a feed-forward neural network model using TensorFlow and TFRecordDataset API, which significantly improved the platform's recommendation capabilities.
• ➢ Continuous rollout and A/B experimentation across different cities, have contributed to the platform's high performance and user satisfaction.

Sep 2017 – Jan 2019 · 1 yr 4 mos · Singapore

• Commission Service:
• ➢ Designed, Implemented and rolled out the cloud-based commission service handling all drivers/merchants commission transactions across all the businesses.
• ➢ Commission service is a low latency, highly resilient service written in Go utilizing concepts of rate limiting, hytrix circuit breakers, retries mechanism and distributed caching.
• ➢ Integrated Commission service with Datadog, Kibana, tracer APIs for monitoring and logging.
• ➢ Added useful libraries for Host-Manager, Distributed Sorted Queues, CoTaskWorkers etc.
• Driver State Manager Service:
• ➢ Driver State Manager (DSM) is a distributed backend service tracks drivers’ online status over a trip life-cycle, location, booking details, taxi-types, preferences and other useful metadata.
• ➢ DSM listens to multiple Kafka streams as data sources, handles the data race gracefully and compute driver states with respect to booking state.
• ➢ DSM is a highly scalable, reliable, fast service heavily used by analytics and allocation teams. DSM provides valuable metrics about driver's usage of Grab platform that has led to build better products.
• Grab is more than just the leading ride-hailing and mobile payments app in Southeast Asia. We use data and technology to improve everything from transportation to payments and logistics across a region of more than 620 million people.

Sep 2016 – Sep 2017 · 1 yr

• Seasoned technologist with expertise in delivering time-to-market solutions for the APAC Equities Trading Business in the Global Market Access field. As part of my role at Citibank Singapore, I specialized in Exchange Connectivity (Execution-side, Asia Pacific), drawing upon my extensive knowledge of key exchanges such as Osaka Exchange, Singapore Fu&Op, SGX Cash, and BATS.
• My technical proficiency spanned a range of programming languages including C++, Java, and Shell scripting, and I was well-versed in a variety of exchange protocols such as OMX/FIX/Binary, Vella (SRLab), 29West UME, and Tibco EMS. Additionally, I had a talent for Application UML Design and Implementation, and was experienced in using Latex.
• As a leader, I managed a midsize engineering team, focusing on growing individual contributors in their roles and ensuring the team's day-to-day work and progress aligned with organizational goals.

Aug 2013 – Sep 2016 · 3 yrs 1 mo

• ➢ As a developer for the Global Market Access Team at Citibank, I spearheaded the development of high-performance libraries and infrastructures that significantly reduced latency overhead and improved market exchange connectivity.
• ➢ I took ownership of the entire development process, from performing unit-testing to performance testing, using our Continuous Integration Framework (CIF) to ensure code quality and maintainability. My extensive knowledge of transport layer integration allowed me to seamlessly integrate our infrastructure with messaging middleware products such as 29West UME and Tibco EMS.
• ➢ In addition, I played a critical role in developing and maintaining core components, including event dispatchers and queuing models, to ensure smooth and efficient communication between different systems. To further enhance our platform's capabilities, I leveraged a wide range of static and dynamic code analysis tools, including Boost Unit-Test, File System Framework, Astyle, and Doxygen.

Apr 2012 – Apr 2013 · 1 yr · India · On-site

• Golpo is a platform to discover the best stories about your favourite places !!