Akash Pandey

Work History

EXL Service Private Limited

01.2024 - Current

Leading customer engagements in the capacity of a project manager cum senior machine learning engineer.
Delivered a workforce forecasting workflow that trains multiple autoregressive and regularized regression models to predict the FTE requirements across loan approval stages for a major American financial services company.
Stewarding the development, and project execution for a GenAI based tool to create IE reports for a renown O&M services provider.
Core member of xtracto.ai, owning the development and improvement of information extraction features.

Canada Revenue Agency

05.2023 - 12.2023

Improving and creating new dev tools/plugins to enable developers to perform integration testing and deployments to ensure quality and compliance.
Created proofs of concept with J2EE to laterally support developers in their daily work to enforce compliance in large-scale mission critical systems.
Optimize the downstream and upstream project dependencies by extracting business logic into multiple repositories.

EXL Service Private Limited

09.2021 - 08.2022

Led a team of data analysts and machine learning engineers, owning the workflow design and maintenance of machine learning APIs integrated with robotic process automation tools.
Research, implementation and providing corporate training globally about deep learning model architectures and how to utilize them at scale.
Liaising with operations to delegate data creation for model development and training. Deployed a suite of models to support the launch of an advanced information extraction tool, which resulted in FTE reduction for a renowned property insurance company by over 15%.
Delivered a SKU optimization focused demand forecasting framework for the largest global supply chain company.

KredX

01.2020 - 09.2021

Reduced the vendor onboarding time by 25% by automating document validation and entity extraction performed by operations supported by multi-purpose deep learning models in production.
Spearhead development of machine learning applications on Azure as well as AWS cloud resulting in expedited delivery and robust ML pipelines.
Collaborated with multi-disciplinary product development teams to identify performance improvement opportunities and integrate trained models.

TransOrg Analytics

11.2018 - 01.2020

Implemented a RFM based CLTV model used for customer profiling.
Implemented customer profile de-duplication in a highly untidy database with the help of graph connected components.
Developed and implemented predictive models to optimize revenue management, demand forecasting, and pricing strategies, achieving a cumulative increase in revenue of $4 million.
Applied NLP techniques to analyze customer feedback and online reviews, enabling the identification of key trends and actionable insights to improve service quality, resulting in a 10% increase in customer satisfaction scores.
Successfully implemented lead scoring models enabling accurate prediction of lead quality and conversion probabilities, resulting in a 13% increase in lead-to-sale conversion rate.

Xoriant Solutions Private Limited

10.2016 - 11.2018

Executed machine learning proof of concepts for leading consulting companies, multinational banks and tech giants.
Developed use cases around real time sentiment analysis, document indexing, document parsing and web application development.
Delivered a VCS change management and release recommendation system for a globally renown networking product company.