Alkesh Chauhan
Engineering Leader at Confluent | Ex-Amazon, AmericanEx, Paytm
13.5
Years of Experience
Education
nsit netaji subhas institute of technology
Companies
confluent, paytm payments bank, iron lady, freelance, paytm, amex, guavus, amazon, amdocs
Reach out to Alkesh Chauhan via Email, InMail and SMS drip
by installing Chrome extension
Alkesh's contact details
Email (Verified)
alkXXXXXXXXXXXXXXXXXXXXXom
Mobile Number
+91XXXXXXXX31
Experience
2021 - Present
confluent
Engineering Manager II
Building a flexible, accurate, performant, and robust monetization platform responsible for supporting flexible pricing and payments, on-demand bill generation allowing Confluent to define and sell products globally.
2020 - 2021
paytm payments bank
Senior Engineering Manager
Delivering solutions and driving development of fast data and realtime ingestion pipelines for Paytm Bank components. Working on solving complex challenges towards serving realtime data needs of multiple use-cases like passbook, statements and rule engines. Hands on with coding/development and working with various stakeholders and helping convert requirements into working tech modules. Managing, mentoring and performance review of engineers. Participating actively in hiring and nurturing of talent.
2020 - Present
iron lady
Guest Speaker
Volunteer and Guest Speaker at IronLady Community to guide and support women in corporate through Art of War Fundamentals and modern day warrior techniques to win at corporate.
2020 - Present
freelance
Author
Children book series to provide gender agnostic upbringing ideas to raise strong new age warrior children. Buy a copy here - www.amazon.com/dp/935765402X
2019 - 2020
paytm
Engineering Manager / Principal Engineer
Created and delivered analytics panel for UnifiedMerchantPanel. Vital business kpi’s like successrate, txn rate etc were displayed on this panel. Payment Gateway Kafka streams were read in Spark Streaming and selected metrics were exported to ES. This has been used to aggregate business critical metrics on query time and for search purposes. 4million merchants are using UMP panel on daily level, with 10million number of queries hitting the ES per day. VMN While Virtual Mobile Number (VMN) project was created to support field agents who need information about their visiting merchants, to look at his attributes to be able to onboard/upgrade/sell insurance or loans. These attributes were being pulled from various sources and we created multiple aggregated datasets to finally merge them together to serve these attributes. 4k fse’s hit this service everyday during business hours. Roles & Responsibilities: ¥ Analysis of requirements with BusinessTeams, HL&LL design and integration of modules. ¥ Development of modules in Kafka, Spark Streaming, ES. BusinessPlanningPerformanceMetrics BPPM This project was directly asked from CEO & CDO, to enable Paytm to make data-driven decisions. Worked closely with Business teams to understand their daily challenges in pulling data and came up with a robust plan to automate the entire process, which involved creating an aggregated layer which can directly be fetched by BAs without being dependent on raw tables & dwh resources availability. For rendering this merchant,txn,user level information on map with dimensions/metrics, we used MapMyIndia Apis. These aggregated facts/olaps were also moved to BigQuery through GoogleCloud to enable querying and also for Dashboarding purposes on GoogleDataStudio. Roles & Responsibilities: ¥ Analysis of requirements with BusinessTeams, HL&LL design ¥ Development of modules in Hive, Spark, ES ¥ Sync Process to sync data from hadoop cluster to S3 and BQ via GCS
2014 - 2018
amex
Senior Engineer - II
Project : Opera Offer Recommendation System Opera uses browser history of American express card users through adobe clickstream data, and other sources tracked through cookies which contain customer level event details. This information is used to find out patterns and similarities between customer-customer, offer-offer and customer-offer. Machine Learning algorithms like collaborative filtering and catboost Boosting Algorithms are utilized to rollup this data and provide recommendations for users at real-time when they login through website/mobile app. The system caters to around 2 million calls/day and churns about 100Tb of data per day. Roles & Responsibilities: • Analysis of requirements, high level design. • Lead low level design and integration of modules. • Development of modules in Hadoop MapReduce, Hive, Spark,Java. • Used Hbase for L2 cache, and Redis for fast lookup. • Used oozie for job scheduling, monitoring and failure recovery. Project : Hubble Graph Database & Telescope Querying Engine Hubble is a platform to provide graphical view over the AmEx transactional database. It provides a graphical view of analyzing the relationships between Customers and Merchants. Telescope layer sits on top of Hubble and provides an interface for users to query Hubble. It enables the users to design ad-hoc queries to Hubble. Implemented Market Basket Analysis to calculate apriori scores for customer-offer similarity. Roles & Responsibilities : • Designed and development using Hbase, Java and MapReduce Jobs using Redis, Solr . • Developed Market Basket Analysis using aprioi algorithm on distributed Hdfs environment
2012 - 2014
guavus
Sr Software Engineer - BigData
Project : Verizon Realtime Analysis Generation of hourly subscriber level report for Verizon Broadband. In earliest deployment scenarios tool was required to work with input data rate of 9 million subscribers, 9kFps IPFIX, 2000 Gateway Routers, 2000 Bulkstats- Interface/qos files. Daily processing of around 4.5billion records /day. Roles & Responsibilities: • Framework Design and analyzing requirements. • High level and low level Design of components. • Implemented SPARK jobs in Java and Scala for modules. Project : Sprint Auto Device Detection Auto-Discovery of network elements for Sprint Communications based on its IP Address. The framework would pick up Sprint data with network input, DNS dumps and merge this data with the standard naming conventions provide by device manufacturers to analyze and compute the device type. Roles & Responsibilities: • Framework design and analyzing requirements. • Implemented HADOOP MAP REDUCE jobs for annotation logic, parsing and lookup • Used OOZIE to schedule these jobs at regular intervals.
2011 - 2012
amazon
Software Development Engineer
Worked as a Software Development Engineer for Endless.com team which focused exclusively on shoes and accessories, now called Amazon Fashion. Implemented highly-reliable, multithreaded and high throughput features. Endless.com was Amazon’s first separate e-commerce brand and it launched sites in Japan, UK, France and Germany. Roles & Responsibilities: • Involved in design and development of backend architecture. • Development using data structures, algorithms and finest analytical skills to reduce latency on site. • Worked on enhancing search experience on the site • Worked on feature enhancements on product pages.
2010 - 2011
amdocs
Subject Matter Expert
Worked as SME-I Developer on designing, developing and debugging/maintenance of software systems in Amdocs
Experience
68 Skills
Algorithms
Algorithms
analytics
apache
Apache Hive
Apache Spark
architecture
architecture
Awk
backend
Backend
Big Data
Big Data
business planning
Cassandra
Core Java
Data Analytics
Data Structures
Debugging
Design
E-commerce
Eclipse
Education
Elastic Search
ElasticSearch
Extract, Transform, Load (ETL)
Google BigQuery
Hadoop
Hadoop
HBase
Hibernate
Hiring
Hive
Hive
Integration
ios
IP
Java
Java
kafka
Kafka
Leadership
Linux
Logstash
Low-Level Design
Machine Learning (ML)
MapReduce
ml
Multithreading
MySQL
Oozie
oozie
Payments
PostgreSQL
Problem Solving
Programming
Python
Recommendation system
Redis
Routers
Scala
Scala
Search
Senior Software Engineer
Shell Scripting
Software Development
Software Engineer
Spark
Education
2006 - 2010
nsit netaji subhas institute of technology
B.E.
Computer Science Engineering
