Mar 2020 – Nov 2021 · 1 yr 8 mos

• Peak is a UK-based startup that is trying to build a new category in the artificial intelligence market which is known as decision intelligence. At Peak, I am part of the solutions team which is responsible for developing AI-based solutions in the cloud.
• Highlights of my work at Peak:-
• Improved the performance of querying on Segment Explorer by 99.9% in real-time by introducing a new analytics database - Apache Druid. Segment refreshing now takes 15 minutes for 160+ segments which were previously taking 8/10 hours previously.
• Developed a fully managed service for time-series forecasting and retail inventory planning. Designed and implemented the fault-tolerant architecture for the entire distributed system. Later, we have organized a 21-day forecast challenge in partnership with AWS which generated 10M+ revenue in the pipeline.
• Demo of AWS Forecast Product: https://www.youtube.com/watch?v=pTzzQXrih9g
• 21 Day Forecast Challenge: https://aws.amazon.com/marketplace/pp/prodview-2cafgl7q74zai
• Implemented Serverless MicroServices using AWS Lambda for several use-cases. It used AWS CloudFormation as an IaC tool along with other DevOps CI/CD best practices.
• Created data ingestion pipeline running on an Apache Spark cluster which can process and ingest more than 5TB of data on daily basis. This also includes validating input data against defined schema rules as part of data processing.
• Wrote code for creating reliable full data lake solution on S3 based on delta lake storage layer.
• Technologies used: AWS, Apache Spark, Java, Node.js, JavaScript, PostgreSQL, Distributed Systems

Jul 2019 – Mar 2020 · 8 mos

• Worked as a Backend Engineer on product Tookan which is a Delivery Management Software for Enterprises.
• Implemented product-delivery scheduling feature for agents and customers. This helps in better scheduling of pickups and deliveries of products. Also implemented cron job for automatically rescheduling incomplete tasks to best possible next day of the week.
• Performed various server optimisations by reimplementing legacy modules, writing optimised database queries. This helps in performance improvements and scaling of the entire system.
• Wrote production level code for features like agent wallet, logging of all dashboard activities for admins, integrating new webhooks etc.
• Technologies used: AWS, Node.js, JavaScript, PostgreSQL, MongoDB

May 2019 – Aug 2019 · 3 mos · Remote

• Worked on adding numerous features and bug fixes in order to release Amahi 12. Implemented new plugins and fixed existing ones including disk-wizard, web-apps generator.
• Also Implemented backend and plugin part of friending feature. This will help Amahi users to share media contents with other Amahi users with proper permissions control.
• Work done during GSoC:
• https://sukhbir-singh.github.io/blog/2019-08-24/gsoc-2019-final-submission-report

Sep 2018 – Feb 2020 · 1 yr 5 mos · Remote

• Member of Ruby on Rails Team of Amahi.

Apr 2018 – Aug 2018 · 4 mos · Remote

• Worked with the core team of Amahi to improve their Ruby on Rails app - Platform.
• Amahi Platform is a web-based app that allows management of users, shares, apps, networking and other settings in a Linux-based system.
• Report on work done during GSoC:
• https://sukhbir-singh.github.io/blog/2018-08-05/gsoc-2018-final-submission-report

May 2017 – Jul 2017 · 2 mos · Mumbai Area, India

• Worked as a Full Stack Developer, added multiple new features such as notifications, profiles, preferences, etc.
• Technologies used: Ruby on Rails, JavaScript, React, MongoDB