Jun 2024 – Sep 2024 · 3 mos

• Took a detour from spreadsheets and models to chase mountain trails, inner peace, and passport stamps. Started off wandering through Northeast India, soaking in nature and adventure, followed by a silent Vipassana retreat (yes, no talking — and survived!). Swapped laptops for handlebars in a mountain biking course, then stuffed a backpack and zigzagged across Europe — meeting old friends, making new ones, and collecting stories (and probably a few bad hostel coffees). Came back recharged, a little muddier, and a lot more inspired.

Jan 2023 – May 2024 · 1 yr 4 mos · Bangalore Urban, Karnataka, India · On-site

• Glance (InMobi) personalization team.
• Functioning as a full‑stack data scientist, involved in all aspects of the project lifecycle, from designing and architecting systems to data engineering, pipeline building, research, model development, deployment, monitoring, and post‑deployment practices.
• Devised an innovative approach to effectively address the cold item problem, leveraging a unique blend of collaborative filtering and similarity matching. This technique resulted in a 12% increase in engagement metrics.
• Worked on ranking engines utilizing multi‑arm bandits and counterfactual evaluations to tackle the cold start problem. Implementing the multinomial Thompson sampling approach led to an 18% improvement in engagement metrics.
• Built recommendation engine for a new form of live content. The model increased the DAU and time spent by 5% & 9%, respectively.
• Built an ML monitoring and evaluation platform to ensure the continuous improvement and optimization of the recommendation systems.

May 2021 – Dec 2022 · 1 yr 7 mos · Bangalore Urban, Karnataka, India · On-site

• Glance (InMobi) personalization team.
• Working in the personalization team to build scalable real-time recommendation systems for Glance's 125 million users.

Jan 2021 – May 2021 · 4 mos · Bangalore Urban, Karnataka, India

• Worked in Support Assist data hub team, which designs applications/pipelines for the SAC telemetry tool for Dell devices.
• Data engineering: All things Azure, Spark, Scala.
• Filed two US patents.

Sep 2020 – Jan 2021 · 4 mos · Gurugram, Haryana, India · Remote

• Built state-of-the-art RL-based recommendation systems for Zomato's 100 million users that support personalization and the sequential nature of user interaction to maximize long-term user engagement.

Jul 2020 – May 2021 · 10 mos · Remote

May 2020 – Jul 2020 · 2 mos · Bengaluru, Karnataka, India

• Worked with actuarial teams to design several classifiers for fraud detection and sequential prediction of the outcome for the different stages of a lifetime mortgages application.
• Developed a system for automatic summarization of dark data of insurance and banking clients without missing or altering any facts and figures.

Mar 2020 – Mar 2020 · 0 mo · Remote

• Collaborated with IBM Research Labs under the Global Remote Mentoring program.
• Worked to develop a system that given a limited number of Gold standard pairs for NLQ and Expected SQL, how can an existing NLIDB system learn from the feedback to answer similar queries in the future.

Dec 2019 – Jan 2020 · 1 mo · Bengaluru, Karnataka · On-site

• Designed statistical pricing techniques using locality dependent elasticity in setting dynamic pricing for hotels on the platform.
• Built visualizations and dashboards using ElasticSearch and Kibana.

Sep 2016 – Nov 2016 · 2 mos · Bengaluru, Karnataka