Vinay Kumar Thanikonda

About Vinay Kumar Thanikonda

Big Data professional passionate about designing highly scalable data processing infrastructure and read optimized petabyte scale datasets for AI/ML and Analytics use cases using modern distributed computing frameworks and data modeling. Firm believer of data driven Business innovation.* Built highly scalable on-demand distributed computing architecture for Amazon offer ranking science team. Decoupled compute and storage on AWS EMR/S3 allowed scientists to run 8X more experiments. Redesigned Big Data architecture saved months of oncall hours for science and engineering teams, improving training and experimental metrics data pipeline run times by 50%.* Built graph clustering based fraud detection ML models using Louvain and SVD algorithms for AWS GroundTruth/Mturk team. Machine Learning(ML) pipeline infrastructure set up is deployed using CloudFormation stack. End to end pipeline orchestration is managed by event-driven notifications. Output from ML models fed into Amazon fraud investigation operational systems reduced false positivity rates by 35%.* Built near real-time AWS GroundTruth\'s micro-services log compaction process using AWS Step Functions and EMR. Data output from this workflow feeds human-in-the-loop CV image labeling quality monitoring systems, detecting bad quality image labelling work submitted by humans instantaneously. This set up enabled team to view a snapshot of CV labeling work by humans in near real-time and act upon bad actors quickly.* Collaborated with 10+ Amazon engineering teams for enabling Amazon Photos Marketing team to have data insights on CPA(cost per new customer acquisition) metrics, influencing 2M$ Marketing budget decision. As part of this effort, surfaced data logging bugs in different engineering systems, worked with software developers to fix data logging problems and enabled fair ranking of Amazon Photos Prime Marketing campaigns on internal Amazon advertising systems. Improved ranking of Marketing campaigns resulted in 100K new customer acquisitions.* Managed Petabyte scale Data warehouses and Data Lakes at Amazon and TESCO* Reduced runtime of TESCO\'s store-product level stock forecasting system by 50% using Teradata\'s Multi-level partition primary index and database optimizations.Technical Expertise - Real time streaming tools: Kinesis Data Streams, Firehose, Managed streaming for Kakfa- Hive, Pig, Spark, Presto,MySQL, Redshift, Teradata, S3- Python, Scala, Java- NoSQL Database : Dynamo DB, Hbase- Container tools: Kubernetes, Docker- Orchestration tools: AWS Step Functions, Airflow

Experience

1. Senior Big Data Engineer-data Science

   Amazon

   Sep 2017 — Present · Seattle, WA, US

   Build scalable cloud data infrastructure platforms for ML scientists*Handle Petabyte scale elastic data storage volumes on Hadoop based workflows using Hive,Pig and Spark*Automate production ML/Statistical models on AWS services

Education

• Maharishi vidya mandir

  High School, SSC

• Andhra University

  Bachelor of Engineering (B.E.)

• Narayana Junior college

  IPE, Mathematics

Skills

• Emr
• Nosql
• Python
• Red Hat Linux
• Shell Scripting
• Data Integration
• Business Intelligence
• Hive
• Database Design
• Extract, Transform, Load (Etl)
• Data Warehouse Architecture
• Unix
• Data Modeling
• Hadoop
• Apache Spark
• Data Warehousing
• Distributed Systems
• Sql
• Databases
• Apache Pig
• Sql Server Integration Services (Ssis)
• Database Optimization
• Ab Initio
• Performance Tuning
• Etl
• Oracle
• Data Mining
• Requirements Analysis
• Teradata
• Parallel Computing
• Big Data Analytics
• Amazon Redshift
• Amazon Web Services (Aws)