Sep 2023 – Dec 2024 · 1 yr 3 mos · Bengaluru, Karnataka, India · Hybrid

• Infrastructure Cost Reduction by 210% by Designing, implementing, and
• maintaining Linux-based infrastructure solutions, including server
• architectures, networking configurations, and storage systems.
• Ensuring the security and compliance of Linux systems by implementing access
• controls, encryption mechanisms, intrusion detection systems, and regular
• security audits.
• Developing automation scripts and orchestrating workflows using tools like Ansible,
• Terraform, GitLab CI/CD and Kubernetes to streamline system pr5ovisioning,
• configuration management, and deployment processes, leading to Organization-
• wide Developer Productivity improved by 3x.

Feb 2023 – Sep 2023 · 7 mos · Bengaluru, Karnataka, India · Hybrid

• SDLC, CI/CD, Networking, System, Consultant
• End-to-End System Design and Architecture Consultation.
• Consulting Clients on Cloud onboarding, migration from monolith to microservices, IoT
• Contributor to Open Source by creating several useful tools and libraries.
• Training clients on BigData, IoT, Cloud, Distributed and Scalable Architecture.
• Working on Medical Project with AI/ML using ChatGPT
• Architected Scalable Question Answering with LLM and RAG

Aug 2020 – Feb 2023 · 2 yrs 6 mos · Bengaluru South, Karnataka, India · Hybrid

• Caterpillar: Scalable Infrastructure(IaC) and CI/CD as Code
• Tools: Linux, Shell, Python, GoLang, Docker, ECS, EC2, Jenkins, EKS (Kubernetes)
• Resposibilities and Impact:
• ∗ Creation of Complete MicroServices Infrastructure and CI/CD as Code
• ∗ Backward compatible to CRED codepipeline v1
• ∗ Completely Developer Driven CI/CD (Self-Serve)
• ∗ Any step can be plugged in the CI/CD pipeline
• ∗ Easily integrated Data Pipelines, MLOps Deployments
• ∗ Currently handling 500+ microservices, 800+ deployments daily
• ∗ Reduced Cost of Release Engineering by 15x
• ∗ Improved Developers, SREs and QA teams productivity by 5x
• RDS-RS-Pipeline: AWS RDS to RedShift Data Pipeline
• Tools: Linux, Python, Docker, ECS, RDS, RS
• Resposibilities and Impact:
• ∗ Scalable and Near Real-time (Micro-Batched) data pipeline
• ∗ Configuration Driven to add more data pipelines
• ∗ Reduced the downtime to 0 (Zero)
• ∗ Newer pipelines can be added dynamically
• CodePipeLine Security: Shift Left Security in CI/CD
• Tools: Linux, Shell, Python, Docker
• Resposibilities and Impact:
• ∗ Plugin based Security stage in CI/CD which catches security related
• issues even before code gets compiled.
• ∗ Git code scanning with rules, regex and entropy
• ∗ Dockerfile linting to apply security best practices
• ∗ Service Bill of Material (SBOM)
• ∗ More plugins can added and enabled dynamically
• ∗ Reduced friction between security and dev/ops teams

Jul 2018 – Sep 2019 · 1 yr 2 mos · Bengaluru, Karnataka, India

• Architect(Systems Engineering) at CRED(https://www.cred.club/) DreamPlug Technologies.
• Complete State-less CI/CD Pipeline
• Designed, Architect-ed, and Developed CI/CD pipeline
• Rollback to any version of code deployed
• Supports micro-services (docker, ECS-EC2, ECS-FarGate)
• Major service related knob controls are provided to Devs to control
• Supports Blue-Green Deployment
• Supports Canary Deployment
• Since this is serverless architecture, this pipeline has facilitated the following features :-
• # release engineering team is not required
• # minimal/zero interaction required between dev and devops
• Mentoring team member to support Batch Jobs(AWS Batch) deployment in this CI/CD pipeline
• One of the First Major users of AWS ECS-FarGate in India at scale
• Golden-Retriever: Secrets+Config injector for micro-services
• Retrieves Secrets/Configs from Secrets store
• Inject those as environment variable in the docker container before starting micro-services
• Helpful in automating Code Deployment in Ci/CD pipeline
• FEKLOg Fluentd ES Kibana +s3 Logging Framework
• Scalable logging framework for micro-services
• Customized fluentd to scale at service level, hence 1 layer of fluentd-servers is eliminated.
• Support archiving through s3, hence supports archived logs through Athena
• Currently, this is able to handle 1.5+ billion events per day. (300+ GB logs per day)
• Lakshman-Rekha:
• Fully automated creation of Security Groups with audit info for PCI-DSS

Sep 2019 – Apr 2020 · 7 mos · Greater Bengaluru Area

• Senior Software Engineer(Data Platform) at Twitter
• Data Pipeline Platform: ETL and Workflow Team
• Tools: Linux, Python, Docker, Google Kubernetes GKE, Airflow
• Resposibilities:
• ∗ Implemented secret/configs injection into the Kubernetes pods of Airflows
• ∗ Adapted Airflow Kerberos SSO with Internal Twitter SSO
• ∗ POC done to deploy Airflow on Kubernetes

Jun 2017 – Jul 2018 · 1 yr 1 mo · Bengaluru, Karnataka, India

• Egress Traffic Platform: Platform to capture, analyse, and control egress
• traffic from Network
• Tools: Linux, Go, Kafka, Redis, ElasticSearch, Kibana, Python
• Resposibilities:
• Network Packet Capture and Refinement at Egress
• Publish data on to self managed Kafka Cluster
• Enrichment of captured data with Reverse DNS lookup data
• Storage of data finally over ElasticSearch cluster
• Analytics of Egress Traffic data using Kibana
• API based firewall traffic control of Egress
• Multi-cloud Infrastructure As Code Deployment Platform: Platform to
• create infrastructure alongwith required softwares deployment.
• Tools: Linux, Chef, Terraform, Shell, Python, Ruby, Git
• Resposibilities: Architecture, Design, Development
• ∗ Gathered requirements from various dependent teams to create the plat-
• form
• ∗ Created POC to show full End-to-End one command/click infrastruc-
• ture(with software) deployment
• ∗ Designed and Created core modules of the Project, like Instance+Chef
• Bootstrap(Rasoiyya), Cookbook Syncer and Recursive Dependency Re-
• solver(Kirana), etc.
• ∗ Mentored other team members to complete other modules
• ∗ Helped other teams/projects to onboard by creating multi-cloud cook-
• books

Mar 2015 – Jun 2017 · 2 yrs 3 mos · Bengaluru Area, India

• Lead Insight Engine: Platform to store and enrich Leads
• Tools: Linux, Apache, MySQL, PHP, Java, Hadoop, MongoDB, Elastic-
• Search, GoogleMaps, Pentaho, Python
• Responsibilities: Architect, Design, Develop
• Web crawling of Online Merchants from Online Marketplaces.
• Importing Leads from partners.
• Enrichment of Leads:- Dedup, Merge, Validation, Auto-Correction.
• Prospective Leads are evaluated based on Data Quality scores.
• Data Pipeline based on ETLs, is created to populate the lead data onto
• centralized repository.
• ∗ Levaraging this repository to Analytics, Marketing, Sales and Underwriting teams.
• ∗ Created Lead generation system by Sales team. [More explanation in Volunteer Section]
• ∗ Importing qualified Leads into CRM System from the repository.
• InnoShop and InnoWow: Heading the team to create Platform to plug Kiranas, FMCG companies, and
• end customers.
• Tools: Linux, Apache, MySQL, PHP, Java, R, Hadoop
• Resposibilities:
• ∗ Understanding the Product Requirement Documents and creation of SRS
• ∗ Creation of the Architecture and Algorithms
• ∗ DB Migration and Backward Compatibility Check
• ∗ Management of Software Testing and Deployment of software.
• ∗ ETLs are written to populate the data onto MySQL
• ∗ BigData Analytics of Kirana Consumer data using R and Hadoop.
• InnoShop and InnoWow:
• Heading the team to create Platform to plug Kiranas, FMCG companies, and end customers.
• Tools: Linux, Apache, MySQL, PHP, Java, R, Hadoop
• Resposibilities:
• ∗ Understanding the Product Requirement Documents and creation of SRS
• ∗ Creation of the Architecture and Algorithms
• ∗ DB Migration and Backward Compatibility Check
• ∗ Management of Software Testing and Deployment of software.
• ∗ ETLs are written to populate the data onto MySQL
• ∗ BigData Analytics of Kirana Consumer data using R and Hadoop

May 2013 – Jul 2014 · 1 yr 2 mos · Greater Indore Area

• Optum Stingray: Configuration Driven Distributed Data Validation and Data
• Filteration Framework
• Client: United Health Group
• Tools: Java, Hadoop, Pig, Oozie, Apache Commons Configuration API
• Resposibilities:
• ∗ Created Security Layer over Configurations to secure sensitive information
• ∗ Created configuration driven distributed data validation and data filtering engine using Hadoop, Pig, Oozie and Apache commons Configuration.
• ∗ Created Distributed FTP to HDFS data transfer framework for data ingress and egress of Hadoop HDFS.
• Neustar NIS: BigData Analytics of Internet Users Segmentation by using URL
• Classification and IP Clustering from ultra DNS logs.
• Client: Neustar
• Tools: Java, Hadoop, Hive, Pig, Apache Nutch, Mahout, R, Shell
• Resposibilities:
• ∗ Participating in design discussions and analyze the problem in detail to
• understand the key points
• ∗ To research and explore technologies, algorithms that help in arriving
• at a solution that is optimal
• ∗ Written ultra DNS parser to use PCAP-ng as Hadoop InputFormat
• ∗ Crawling and Fetching URL contents using Apache Nutch
• ∗ Found and Fixed OOM issue in ParseSegment Phase of Apache Nutch
• (Jira: NUTCH-1640)
• ∗ URL categorization based on client requirements.
• ∗ IP clustering based on the above categorization using Mahout K-means
• algorithm.
• ∗ Created Hive UDF that maps IP to countries.
• ∗ Used R for initial data analysis and statistics generation
• Created Shell Script Snapshotting Framework for Checkpointing the Full Workflow
• Programming (in C/C++, Perl, Java, Python), and scripting (in Shell, awk, sed, Perl)
• BigData Practice: Project Development with Hadoop and its related softwares like Pig, Oozie. Hive, HCat etc.

Jul 2007 – May 2013 · 5 yrs 10 mos · Bangalore Area, India

• Cloud Platform Group
• Hadoop: HDFS and Map-Reduce framework, Pig, Oozie, Hive, HCat
• Working with customers on their grid application optimization and providing efficient solutions
• Contributor to Open Source by raising many features and defects in Pig, Hadoop and Oozie.
• Training other teams on Hadoop and its components
• Onboarding different applications onto grid
• Managing applications running across 13 clusters, 40,000 nodes.
• Found and provided fix for Race condition in Hadoop 0.20.204
• Pig Scripts for Distributed Data Mining :e.g. MultiLineLoader() UDF
• Found deadlock in hadoop scheduling due to oozie launcher
• Splittable LZO Compression support on Hadoop, Pig, Oozie using Hadoop Distributed-Cache
• Found and fixed Race condition in Giraph-442 ZooKeeper Server
• HDFS Space Saver by analyzing user jobs.
• UDB Crawl Team
• Working with customers for data mining on User DataBase.
• Plugin based framework for UDB Crawl. (Written in C++)
• Automation of Crawler disk usage reporting.
• Integration of Zlib into existing crawl code to reduce disk space requirement.
• Optimization of existing code base for crawl
• Automation of crawl output delivery
• Nagios Network and System Monitoring Plugins
• Hands on realtime issues like DOS attacks, network/system issues
• Automation of YCC user account sync.
• Programming (in C/C++, Perl, Java, Python), and scripting (in Shell, awk, sed, Perl) for User Database Crawl.
• Grid Solutions: Hadoop and its related softwares like Pig, Oozie. Hive, HCat etc.