mayank parashar

Senior Software Engineer

Bengaluru, Karnataka, India6 yrs 9 mos experience

Most Likely To Switch

Key Highlights

Developed a fraud detection service processing over 50,000 documents daily.
Engineered solutions to optimize performance and security.
Strong background in data structures and algorithms.

Stackforce AI infers this person is a Fintech and SaaS-focused Software Engineer with expertise in distributed systems.

Contact

mayank.parashar@moengage.com LinkedIn

Skills

Core Skills

Distributed SystemsPython

Other Skills

Algorithm DesignAmazon Web Services (AWS)Apache KafkaAsynchronous ProcessingCloud ComputingCore JavaData StructuresDjangoDjango REST FrameworkDocker ProductsFSMFlaskGitGo (Programming Language)Golang

About

As a Software Engineer at Ocrolus, I developed and designed tampered document detection service that helps banks and lenders prevent fraud and reduce bad loans. This service processes over 50,000 documents in peak hours and we have processed more than 25 million documents till now. I have expertise in designing, developing, and deploying scalable and reliable distributed systems, using technologies such as Python, Java, Golang. I have also contributed to the optimization and improvement of the service's performance, usability, and security, by implementing features such as a generic template builder, an asynchronous processing flow, and a solution to mitigate race conditions. I have four years of experience in software engineering, working on various projects and domains, such as cab service, policy engine, and enrichment service. I have a strong background in data structures and algorithms, and I enjoy solving complex and challenging problems. I graduated with a Bachelor of Technology (B.Tech.) in Computer Science from Dr. A. P. J. Abdul Kalam Technical University in 2019, where I won multiple coding contests and was a finalist in a national level competition.

Experience

Moengage

Senior Software Engineer

Oct 2023 – Present · 2 yrs 5 mos

Built an automated email unsubscription feature, managed user opt-outs while handling rate limit exhaustion with retry strategies and timely failure alerts to ensure proactive issue resolution
Reduced the cost of list unsubscribe happening because of storage of links for redirection innovating storage less redirection technique
Designed bot click detection service which handling 12000/sec requests which mark click open events as suspicious bot activity

Ocrolus

Software Engineer

Nov 2021 – Jul 2023 · 1 yr 8 mos · Gurugram, Haryana, India

Build tampered document detection service from scratch, which process different kind of documents and detect fraudulent loan applications.This helps banks and lenders to save money on bad loans and decrease NPA.Currently this service handles over 50k documents per day.
Developed a generic template builder identify paystub origin using a YAML configuration, which simplified the onboarding process for new templates reducing TAT to add new template from hours to few minutes and skip 10% of the total paystubs
Redesigned the Service to run sources Asynchronous to make it instant, helps to reduce our average result wait time from 5 mins to less than 15s
Engineered solution to mitigate race conditions occurring after lock release(due to timeout) in a distributed service by utilising low level signals to ensure that process terminated gracefully, eliminating potential background processing issues.

PythonJavaGolangDistributed SystemsDjangoAsynchronous Processing

Cyware

Software Engineer

Nov 2020 – Oct 2021 · 11 mos · Bengaluru, Karnataka, India

Design optimal async flow between policy engine, enrichment service and confidence score microservices. use batch processing and async polling between services to minimise resource utilisation and eliminating continuous socket connection.

Hashedin

2 roles

Software Engineer

Jul 2019 – Nov 2020 · 1 yr 4 mos · Bangalore

Glyd/TreoMahindra: A cab service based on electric transport for small distance. Develop whole Trip flow with using FSM.
Development and optimisation of a scalable Booking Engine for finding best ride for
Utilisation of Node.js for high I/O Bound process like Location service, notifications