* Developed proprietary NLP and Search algorithms for Talk With Me App by NZ basedInclusys. 

Talk With Me is an educational resource for educators and parents to help children with autismlearn social interaction skills. The development of the app is supported by CallaghanInnovation, New Zealand Government Crown Entity and KiwiNet, that supportscommercialization of publicly funded NZ research.

Senior Software Consultant

Developed React components and CI/CD pipeline for Labxchange at Open edX.

Worked in information retrieval, search relevancy and ranking and machine learning domains to build India's first doubt resolution platform for K12 students - Gradeup School, featured in Google IO 2019

* Setup the whole text and image search from scratch. Developed custom relevancy and precision control pipeline, which filters negative results thereby making results more accurate
* Worked on Image processing and segmentation using Opencv to detect and segment important part in an image
* Setup a search logging infrastructure and visualization dashboard to track the search and accuracy over certain parameters
* Setup and integrated Rule Engine to give control to the product, business, and market team to make changes and fine tune params on the run-time, thus reducing their dependency on the engineering team
* Setup a Task runner and execution interface, which reduces dependency on the engineering team for ad-hoc tasks, and give power to product and operations team to quickly get their tasks done
* Setup and promote open source and blogging culture

Gradeup, a Times Internet-backed venture, is India’s largest and most engaged community for online learning.

Senior Software Engineer

Mentored Evalai under GSOC 2017. Onboarded new contributors, reviewed pull requests,mentored contributors and maintained the whole codebase. 

Google Summer of Code 2017, Organization Mentor

Recharges Team

I was responsible to enable and ensure smooth functioning of Business and Operations Team. Projects:

* Developed the Recharge MIS Interface. This helps Business and Operations to track the status of any operator or gateway

* Developed the Live Recharge Status Dashboard to track success/error percentage of any operator, gateway. This helps Business Team in monitoring and deciding the re-routing of any operator through other gateways

* Created the Operator Re-routing Service for Recharges. This enables Operations to re-route any operator at run-time during peak hours or in a case when any gateway is down

Logistics/Fulfillment Team

I was responsible for developing systems to facilitate onboarding of various shippers, handling scale and reducing error rates of the systems when interacting with Shippers.
Projects:

* Worked on reducing New Shipper Onboarding, Integration and go-live time. Developed a customizable, easy to plug module for shippers (plugin) to integrate for services like Order Creation, Cancellation, and Tracking. (https://github.com/paytm/courier-plugin-sdk-nodejs)

* Developed the Shipper Selection and Allocation Simulation Interface for Logistics. This enables Business Team in decision making for giving priority to certain shippers and monitoring the allocation of a shipper for any product along the available routes

* Developed a Product Liquidation Decision Service. It helps in deciding whether the product should be liquidated or paid out to the merchant. This dashboard is being used by various third party Reverse Processing Centers all over India

* Worked on reducing the error rates for the systems when interacting with third-party Shippers. Designed and developed a Generic Retry Service. The service uses an exponential backoff formula to calculate the next retry time

* Implemented the consumer side of Paytm Oauth as a reusable Django application (https://github.com/paytm/django-paytm-oauth) 

Software Engineer

* Created several open packages in Python.

* Used Python for scrapping and Natural Language Processing Tasks.


* Used Django to create a popular open source alternate to Kaggle which is being used by Stanford, CMU, Google and Facebook.

* Published several Django packages.

* Contributed to various open source Django projects by notable organizations like PythonIndia and Mozilla.

Used Elasticsearch to create award winning and impactful search engines.
* Created Image based Search Engine for High School students in India, featured at Google IO 2019.
* Created Grammar based Search Engine for children with Autism while working with Inclusys, which is a New Zealand based startup backed by Callaghan Innovation, New Zealand Government Crown Entity

* Designed and managed database for a popular open-source project.

Used RabbitMQ to create a parallelized backend for an open-source Kaggle alternative. This reduced the submission processing time to 2 minutes from the previous 13 minutes taken by Microsoft's Codalabs.

* Deployed and managed several production-level services using AWS EC2 services.
* Familiar with S3 and RDS.

* Setup database for several production-level applications.
* Managed and optimized database.
* Identified and tuned queries from the slow query log.

You always fear from what you don't understand

Taranjeet Singh

EvalAI is an open source platform for evaluating and comparing machine learning (ML) and artificial intelligence (AI) algorithms at scale.

The project (700+ stars, 350+ forks on Github) is used by organizations like Stanford University, Carnegie Mellon University, Google AI Research, Facebook AI Research. A company is also built around a forked version of it.

EvalAI - A Fast Open Source Alternate to Kaggle and Microsoft Codalab

* Dropbox's Zulip: Refactored code linting pipeline to make it more efficient. 
* PythonIndia's Wye: Created dashboard for tutors to manage their workshops and students. 
* Mozilla's Fjord: Improved test cases for the authentication module. 

Personal details

Taranjeet S.

Summary

Technical skills

Work Experience

Personal Projects