Jan 2022 – Jun 2024 · 2 yrs 5 mos

• Successfully led the launch and scale-up of Flexible Slots in India, resulting in significant impact with partners receiving 6-8 jobs per day and a fill rate of ∼70%. Oversaw the design, planning, and execution of the project, coordinated with multiple teams & developed various capabilities including horizontal scaling of workflows based on load.
• Leveraged flexible slots to launch milk-run model for societies thereby reducing the cost for AC service
• Designed and implemented batching system to optimize customer-partner matches, published a tech blog for the same
• Built the flow for capturing partner feedback on not accepting a job, gathered more than 1M data points
• Collaborated with the growth team to launch the instant bookings pilot, developed custom requirements in matchmaking for successful launch in the US
• Introduced new process for code reviews and organized knowledge sharing sessions within the matchmaking pod of 8 folks

Nov 2020 – Jan 2022 · 1 yr 2 mos

• Matchmaking, Allocation team
• Overview
• Process of finding the best-suited partner to serve a customer's request.
• It involves calculating complex predictive metrics using multiple signals around partner availability, skills, current inventory, preferences, location, historical behavior, and other parameters.
• We need to make constant trade-offs between partner economics, marketplace economics, and customer delight while maintaining the overall fairness of the system.
• Implemented a workflow to sync matchmaking configuration files between local storage & google drive
• The configs were automatically uploaded, downloaded & categorized as per requirement.
• This helped the product managers & business folks to cut all their manual efforts by 60%.
• Built dashboards for visualizing & comparing matchmaking configuration files.
• This helped improve the overall transparency of the matchmaking process owing to the high complexity of the system.
• Worked on the end-to-end deprecation of matchmaking's monolith
• Migration of core matchmaking flows to microservices. Removed external dependencies & reduced the overall response time by more than 50%.
• Was responsible for redesigning, migrating, and platforming many fulfillment use cases. Migrated 46 APIs, 12 Kafka events & backfilled 8 database collections.
• This drastically helped to reduce the tech debt. It also improved the overall system performance since the new system scaled well during peak times and had no downtime, unlike the older system.

Jun 2019 – Jul 2019 · 1 mo · Tokyo, Japan

• Worked on Capacity Planning of CPU usage and analysis of Apache web access logs.
• Created a Logstash pipeline to parse millions of data records & log file sizes in GBs. Used InfluxDB & Elasticsearch to store and index the parsed data.
• Performed exploratory data analysis using Kibana & Grafana to find out the trends, seasonalities over different intervals of time.
• Developed Time Series Forecasting, Anomaly detection models using Prophet & created machine learning jobs in Elasticsearch.
• Created a standalone web application for the same using voila, jupyter notebook & ipywidgets.