Sep 2022 – Jan 2024 · 1 yr 4 mos

• Improved search indexing infra efficiency by moving to an event-driven indexing system.
• For the last eight months, I was a founding member of the ads team at Faire where I helped launch ads and saw it grow from 0->$XM ARR. I led ads indexing and retrieval from the eng side.

Sep 2021 – Sep 2022 · 1 yr

• I am part of the search experience team at Faire where I work on all aspects of the stack - backend, frontend, and also data. I collaborate closely with product and design in coming up with delightful search experiences for retailers.
• I was the tech lead for multiple high-impact (x% GMV lift) search features related to searching for brands and location-based searching for European and Australian users.
• In the past, I was on the checkout team. I specifically worked on product features related to the cart and the product description page.
• Apart from leading big projects from start to finish, I have also helped grow and mentor 5+ engineers at Faire.
• Tools used: Kotlin, Elasticsearch, Airflow, React, Typescript

Sep 2019 – Sep 2021 · 2 yrs · Edmonton, Alberta, Canada

• Area of research: Approximation Algorithms (Theoretical Computer Science).
• Solved a long-standing open problem on the approximability of capacitated vehicle routing problem (CVRP) in various graph classes with Prof. Mohammad Salavatipour. CVRP is a generalization of the travelling salesman problem (TSP).
• Paper related to this work can be found in this link - https://arxiv.org/abs/2106.15034. It was accepted to SODA 2022. Symposium of Discrete Algorithms (SODA) is one of the top conferences for theoretical computer science.
• Link to thesis - https://doi.org/10.7939/r3-ved5-js37

Sep 2018 – Dec 2018 · 3 mos · Waterloo, Ontario, Canada

• Worked with Prof. Shalev Ben-David in the field of computational complexity. The goal of our project was to prove a composition theorem for bounded-error randomized query complexity for partial functions.

Jan 2018 – Aug 2018 · 7 mos · Waterloo, Ontario, Canada

• Worked with Prof. Naomi Nishimura in the field of combinatorial reconfiguration. We worked on two projects:
• 1. Finding an algorithm for reconfiguration of independent sets of even-hole-free
• graphs.
• 2. We defined a new model of reconfiguration where one could move multiple tokens in parallel in one step, generalizing the standard token sliding model. I worked on proving whether or not the model gave an advantage over the existing models.
• Manuscript of project 2 can be found https://adijp.github.io/URA.pdf

Aug 2016 – Dec 2016 · 4 mos · Toronto, Canada Area

• I focused on helping the organization be more data-driven by extracting valuable insight from existing data and using it efficiently to build a smarter organization.
• a) Built an analytics dashboard using c3.js and backbone.js to visualize trends and patterns in causes of Hubdoc’s infrastructure’s downtime. Insights from this dashboard were used for identifying high-impact projects.
• b) Optimized the customer support pipeline by building a machine learning model (in scikit-learn) to automate categorization of customer support tickets using pre-existing support tickets as training data.
• c) Reduced API latency time by over 50% by revamping the Postgres schema and scheduling batch jobs.
• d)Added custom webscrapers and performed maintenance on Hubdoc’s webscraping infrastucture (node.js and phanton.js) which scraped documents for thousands of customers from over 600 vendors worldwide.

May 2015 – Aug 2015 · 3 mos · Toronto, Ontario, Canada

• Performed rigorous testing by creating detailed, well-structured unit and integration tests for OpenText FirstClass,
• a collaborative web application for teams.
• Added accessibility features to the application and made it WCAG complaint (web accessibility guideline).
• Improved the application’s latency by identifying performance bottlenecks in the application and alerting the development team.