Florin Marin

Skills

OS: Linux, Windows
Build Tools: MSBuild, Maven
CI/CD: Jenkins(Including Pipelines and Shared Libraries), GitLab CI, GitHub Actions
Virtualization: VirtualBox+Vagrant, libvirt + Qemu
Containers: Docker
Clustering: Docker Swarm, Kubernetes, Helm
Monitoring: Prometheus and Grafana
Provisioning: Terraform
CMS: Ansible
Foundational technologies and knowledge: Git, Agile, ITIL

Work Experience

DevOps Technical Lead, Ericsson, 2021 - present

Ericsson is one of the leading providers of Information and Communication Technology (ICT) to service providers. We enable the full value of connectivity by creating game-changing technology and services that are easy to use, adapt, and scale, making our customers successful in a fully connected world. Cloud-native is the key to unlocking the full revenue potential of 5G applications. Being first to the cloud means being first to greater scale, interoperability, and efficiency than ever before. 

As a Technical Lead, I aim to communicate with the management and establish a clear 

direction for Infrastructure Development. With a team of 12 people, some of my 

responsibilities are: 

- Create a high performant team, focusing on delivering high-quality solutions that will be 

used on all projects at an organizational level. 

- Provide direction and feedback, establish a roadmap and tasks 

- Adopt, Refactor, Propose, and Implement Cloud Native Solutions 

- Write custom Prometheus Exporters, Queries, and Grafana Dashboards 

- Create custom alerts using Prometheus and Alertmanager 

- Adopt Zero Trust Network 

- Add SSO and LDAP integration to web-facing applications 

- Create a backlog, train juniors, and discuss priorities 

- Refactor the CICD process and add GitOps to the process 

- Automate SSL management using Cert-Manager 

- Test Infrastructure Availability, Reliability, and Observability using Chaos Engineering 

- Update custom Kubernetes Operators to fit our needs 

- Integrate Kubernetes with a private Oauth Server.

DevOps Specialist, Coya Insurance, 2021 - 2021

Coya is a German insurance from Berlin with a completely digital sign-up process  

While working at Coya, some of my responsibilities were: 

- Upgrade Terraform and help with Terraform State surgery. The Terraform Upgrade process is seamless without infrastructure changes. 

- Reduce AWS Costs by adding Reserved Instances, Spot-Instances, and Cluster-Autoscaler. Deploy a fleet of spot instances on ASG. 

- Use 100% spot instances for Dev and Staging Environment. Use saving plans to reduce 

costs. 

- Migrate the existing database to a fault-tolerance cluster using AWS DMS with CDC, 

Aurora Postgres and AWS RDS Postgres and mitigate single-point of failure. 

- Improve Observability by Adding custom Grafana Dashboards, Loki for Logs, and 

creating unified dashboards with logs and metrics 

- Work on backup/restore and disaster recovery plans. 

- Implement AWS SSO and multi-account strategy 

- Help with production downtime, limit PostgreSQL queries, report slow queries, and profile them. 

- Migrate to Graviton Instances for better performance and costs. Migrate from AMD to ARM. Add Docker Buildx and add support for multi-arch build and images. 

- Improve security by removing the need for SSH Keys. Deploy AWS SSM and document it so Developers can have access to databases without VPN, change the Security Groups, use bastions or SSH keys 

- Migrate from Kustomize to Helm Charts 

- Migrate from the Nginx Ingress Controller to Kong 

- Transform logs using Promtail and determine timestamp, level, and other labels. 

- Add alerts to signal system failures and create custom dashboards using Grafana, Prometheus, and Loki to reflect the alerts and system state. 

- Apply best practices to dashboards, like RED/USE patterns, use hierarchy, understand the audience, and avoid the wall of graphs. 

- Migrate from Kustomize to Helm Charts 

- Create Grafana re-usable templates for alerts/notifications 

- Improve crushing Prometheus instance due to slow queries 

- Debug Prometheus query and improve query performance

DevOps Engineer, Siemens, 2020 - 2023

As Cloud Architect/DevOps, some of my responsibilities are but are not limited to:  

- Work on AWS cost optimization, reducing the yearly cloud usage by 300,000 euros. Right- 

sizing the RDS instances, switching to Spot Instances, autoscaling the Development dynamic environments per requests, reducing the unused RDS storage, and switching to RDS Serverless for Dev instances. 

- Implement and architect a complete observability solution using Prometheus stack and other Prometheus Exporters. Configure metrics for each exporter to ensure the metrics' cardinality and a performant Prometheus cluster. 

- Create Grafana dashboards and alerts that apply to general services, applications, or custom implementations using a CICD system and GitOps, following common patterns like RED, USE ( the 4 golden signals). 

- Application instrumentation - create custom metrics for custom applications in Python,  Java and Golang, considering OpenTelemtry, Prometheus conventions, and best practices. 

- Enable applications autoscaling by adding VPA, PDB, and Keda with HPA on custom  metrics 

- Improve Postgres Performance by reviewing slow queries, long-running queries, logs, and  adding indexes. Find queries impacting the CPU, optimize those queries using EXPLAIN, and add proper indexes, reducing the CPU Utilization for the Production Cluster from 100% to 30% during working hours. 

- Demo PoCs and solutions to large company meetings 

- Improve EKS Security by implementing CIS Benchmark and building EKS CIS Images using Packer. Always building on top of the latest AMI. 

- Improve tools and processes toward GitOps and SecureOps 

- Work on production incidents and ensure 99.99 availability. 

- Define alert strategy and incident response strategy at the company level 

- Troubleshoot EKS issues, addons, Kubelet, vpc-cni, fix production issues, clusters stability issues, nodes flapping between NodeNotReady and NodeReady, incompatibility between EKS and self-managed addons (kube-proxy, VPC-CNI)

Technical Lead, DevOps, ShareSafe, 2016 - 2020 

Responsibilities: 

- Develop Level 3 (Maturity Model) REST APIs. 

- Evaluate and recommend new technologies and tools to solve emergent challenges. 

- Make critical decisions and recommendations on varying scope and complexity of technical projects. 

- Turn ideas or simplistic specifications into full-fledged product features, including unit and end-to-end tests. 

- Work with a wide range of audiences, including developers, product managers, customers, sales, and QA, and validate customer needs to turn ideas or simplistic specifications into full-fledged product features to deliver the right solution. 

- Support pre-sales through product evaluations, gathering requirements, building and presenting demos(Proof of Concepts), and defining the scope of work. 

- Develop high-quality, clean, maintainable code in collaboration with other developers using industry-standard best practices, including unit testing and code reviews. 

- Maintain, debug, and optimize scalability, performance, and quality. 

- Migrate identity and access management out of a monolithic application. 

- Support internal shift from a monolithic architecture to a micro-service-oriented model to help scale the business. 

- Design and implement distributed systems architecture based on event sourcing and concurrent programming models. 

- Learn new open-source tools to help get the task done. 

- Responsible for deployments to production, troubleshooting, and helping to resolve customer issues on production deployments. 

- Develop comprehensive monitoring solutions to provide complete visibility and Monitor critical infrastructure systems to avoid errors proactively in development, validation, and release activities. 

- Extend Kubernetes with custom operators (controllers, CRD, Kubebuilder) 

- Write Helm Charts for custom applications.

‍

Stack: RabbitMq, MariaDB, Elasticsearch, Redis, Gitlab CI, Docker, Kubernetes, Helm, Prometheus, Graylog, Ubuntu, GCP, Lua, NodeJS, Golang, PHP, Java, Spring. 

Senior Software Engineer, PayU, 2016 - 2016

Working on integrations of new acquirers, solutions for online credit or instant loans, installments, and other local payment methods. Develop new features in the PayU CEE online payment platform according to business requirements. 

As a member of the Scrum team, my main responsibilities were: 

Developing features for platforms in Romania, Turkey, Russia, Ukraine, and Hungary, which need support in using or implementing the features we develop for them. 

Stack: PHP, MySQL, RabbitMQ, JavaScript, Angular. (SOA) 

Summary of Responsibilities: 

- Work effectively with big databases. 

- Refactor platform code. 

- Build Rest APIs that merchants can integrate with. 

- Write database tests, integration tests, and unit tests in PHPUnit. 

- Develop new payment methods. 

- Code review, appraise and advise on other engineer's code.

Software Developer, null, 2013 - 2016

As a junior/middle developer, I have worked for several international customers (US-based
mostly) on different projects, often as a Full Stack Engineer

Skills: Angular, Jquery, ExtJS, PHP, Codeigniter, Laravel, Docker, MySQL, Jenkins,

Responsibilities:
- Develop frontend and backend
- Manage and implement the SDLC
- Identify the business requirements and implement the solution
- Participate in code review for the release branch
- Manage the LAMP servers and AWS servers
- Write integration tests, unit tests using PHPUnit, and functional tests in Behat