Andrey Kononov, Developer in Batumi, Adjara, Georgia
Andrey is available for hire
Hire Andrey

Andrey Kononov

Bio

Andrey is an infrastructure and software architect with 15+ years of experience, including CTO roles. He now focuses on AI training and fast inference systems. He builds cloud-agnostic, high-performance platforms on AWS, GCP, Kubernetes, and SageMaker for model training, inference fine-tuning, and GPU work. Andrey also designs Python microservices with strong IaC, DevSecOps, and scaling practices. He delivers reliable, cost-efficient AI pipelines that support real business needs.

Portfolio

Geltek
You Only Look Once (YOLO), Kubernetes, Amazon Web Services (AWS), Jupyter...
360dialog
Google Cloud Platform (GCP), SQL, Terraform, Python, Team Leadership...
Kubient (NASDAQ: KBNT)
Amazon Web Services (AWS), PagerDuty, Datadog, Drone CI...

Experience

  • Infrastructure as Code (IaC) - 12 years
  • Rational Software Architecture - 10 years
  • Infrastructure Architecture - 8 years
  • Amazon Web Services (AWS) - 8 years
  • Kubernetes - 7 years
  • ETL - 6 years
  • Google Cloud Platform (GCP) - 6 years
  • Data Inference - 5 years

Preferred Environment

Python, Cloud, Google Cloud Platform (GCP), Amazon Web Services (AWS), IT Projects, Startups, Full-stack, Microservices Architecture, Cloud Architecture, Big Data

The most amazing...

...thing I built is end-to-end AI: GPU infra, fast inference, Python microservices, YOLO skin-detection, and a CVAT tagging pipeline for continuous model learning.

Work Experience

CTO

2023 - PRESENT
Geltek
  • Developed an API that detects skin anomalies with a 95% accuracy rate. Set up the tagging process for contractors using CVAT.
  • Tracked everything in the GitHub ecosystem, set up all development and model delivery processes, patented the technology, and registered it.
  • Launched the app and scaled it to be able to make 100 inferences.
Technologies: You Only Look Once (YOLO), Kubernetes, Amazon Web Services (AWS), Jupyter, gpu operator, Serverless GPUs, Consensus Algorithms, AI Learning, Rational Software Architecture, Load Balancers, Infrastructure Architecture, EMR, ELK (Elastic Stack), Google Cloud Dataproc, Spark, AWS Cloud Architecture, Agile, Linux Administration, Web Security, Data Inference, Team Management, Monitoring, Cloudflare, Rust, Amazon RDS, GitHub, Orchestration, Grafana, Containerization, Git, Visual Studio Code (VS Code), IntelliJ IDEA, GitHub Actions, SOC 2, Firebase

Head of Infrastructure

2021 - 2023
360dialog
  • Led the company transformation and enhanced security. Introduced monitoring tools and led cloud cost tracking and enhancement initiatives.
  • Set up processes for tracking SRE/DevOps efficiency, Enabled continuous security processes.
  • Launched new products and led infrastructure to successfully meet the load (tens of thousands of transactions per second).
Technologies: Google Cloud Platform (GCP), SQL, Terraform, Python, Team Leadership, Leadership, System Administration, Serverless, SQLBase, PostgreSQL, DevOps, VPN, Kubernetes, Go, CI/CD Pipelines, Redis, Java, Prometheus, Amazon EKS, NAT, Infrastructure as Code (IaC), Analytics, Machine Learning Operations (MLOps), Docker, APIs, Cloud Architecture, AlloyDB, GCP Network, Amazon S3 (AWS S3), Amazon SageMaker, FastAPI, Helm, Argo CD, GitOps, LAMP, AWS Lambda, Amazon SageMaker Pipelines, AI Learning, Rational Software Architecture, Ansible, Load Balancers, Infrastructure Architecture, EMR, ELK (Elastic Stack), Google Cloud Dataproc, Spark, AWS Cloud Architecture, Agile, Linux Administration, Google Kubernetes Engine (GKE), Pulumi, Web Security, Data Inference, Team Management, Interconnected Networks, Monitoring, Cloudflare, Amazon RDS, AWS Auto Scaling, GitHub, Orchestration, Grafana, Containerization, Git, Google Cloud SQL, Visual Studio Code (VS Code), GitHub Actions, SOC 2, Firebase, Amazon Elastic Container Service (ECS), Amazon Elastic Container Registry (ECR)

Deputy CTO | Infrastructure Architect

2018 - 2022
Kubient (NASDAQ: KBNT)
  • Migrated all infrastructure from AWS to GCP using Terraform.
  • Led migration activities from AWS SaaS like DynamoDB and Redshift to MongoDB, CockroachDB, and Bigtable.
  • Enable deployment and scaling automation with Ansible.
  • Enabled CI/CD for all internal software with Drone CI.
  • Enabled autoscaling using Datadog, PagerDuty, and Terraform scripts.
  • Tuned a data pipeline for reporting and billing.
  • Introduced infrastructure cost monitoring, which helps reduce the infrastructure cost by three times.
Technologies: Amazon Web Services (AWS), PagerDuty, Datadog, Drone CI, Continuous Delivery (CD), Continuous Integration (CI), Ansible, BigTable, MongoDB, Redshift, Amazon DynamoDB, Erlang, PHP, CockroachDB, Terraform, Google Cloud, SQL, MariaDB, MySQL, System Administration, Serverless, SQLBase, PostgreSQL, DevOps, VPN, Kubernetes, CI/CD Pipelines, Redis, Java, Prometheus, Apache Kafka, Google Workspace, Amazon EKS, NAT, Infrastructure as Code (IaC), Analytics, API Integration, Machine Learning Operations (MLOps), Docker, APIs, Cloud Architecture, AlloyDB, GCP Network, Cloud Migration, Amazon S3 (AWS S3), FastAPI, Helm, GitOps, LAMP, AWS Lambda, Rational Software Architecture, Load Balancers, Infrastructure Architecture, Google BigQuery, EMR, ELK (Elastic Stack), Google Cloud Dataproc, Digital Advertising, Spark, AWS Cloud Architecture, Agile, Linux Administration, Google Kubernetes Engine (GKE), Pulumi, Web Security, AWS Cloud Computing Services, Team Management, Interconnected Networks, Monitoring, Cloudflare, Solidity, GitHub, Orchestration, Grafana, Containerization, Git, Google Cloud SQL, Visual Studio Code (VS Code), GitHub Actions, SOC 2, Firebase, Amazon Elastic Container Service (ECS), Amazon Elastic Container Registry (ECR)

Infrastructure Architect | BidSwitch Operations Team Lead

2017 - 2018
IPONWEB
  • Led activities with the migration to Kubernetes environments.
  • Maintained the system to be cloud-agnostic for having the possibility to choose the best cost options.
  • Improved delivery pipelines and made some serverless systems.
  • Investigated features around AWS and Google Cloud Engine.
  • Led the team of nine talented DevOps engineers and grew the team.
  • Set up and maintained Elasticsearch NoSQL database with a focus on reads.
Technologies: Amazon Web Services (AWS), Elasticsearch, Google Cloud Engine, AWS CloudFormation, Terraform, Puppet, Google Cloud Platform (GCP), Kubernetes, SQL, MariaDB, MySQL, System Administration, Serverless, SQLBase, PostgreSQL, DevOps, VPN, OpenStack, Go, CI/CD Pipelines, Redis, Java, Prometheus, Apache Kafka, NGINX, Email, Google Workspace, Emailers, NAT, Infrastructure as Code (IaC), Analytics, API Integration, Docker, APIs, Cloud Architecture, AlloyDB, GCP Network, Cloud Migration, Amazon S3 (AWS S3), FastAPI, LAMP, AWS Lambda, Rational Software Architecture, Bash Script, DNS Servers, Load Balancers, Infrastructure Architecture, Google BigQuery, EMR, ELK (Elastic Stack), Google Cloud Dataproc, Digital Advertising, Spark, AWS Cloud Architecture, Agile, Digital Transformation, Linux Administration, Web Security, AWS Cloud Computing Services, Team Management, Monitoring, Cloudflare, Amazon RDS, AWS Auto Scaling, GitHub, Orchestration, Grafana, Containerization, Git, Visual Studio Code (VS Code), IntelliJ IDEA, Amazon Elastic Container Registry (ECR)

Senior Solution Architect

2016 - 2017
EPAM systems
  • Participated in pre-sale activities including proofreading materials for customers.
  • Prepared tech strategies for the cloud enablement process.
  • Implemented the project with a completely serverless design.
  • Improved CI/CD pipelines for a large enterprise customer.
  • Researched ways to decrease cloud costs and enabled cost monitoring.
Technologies: Migration, Cloud, SQL, MariaDB, MySQL, System Administration, Serverless, SQLBase, PostgreSQL, DevOps, VPN, OpenStack, CI/CD Pipelines, Redis, Java, Prometheus, Apache Kafka, NGINX, Office 365, Infrastructure as Code (IaC), Analytics, API Integration, Docker, APIs, Cloud Architecture, GCP Network, Cloud Migration, Amazon S3 (AWS S3), FastAPI, LAMP, Rational Software Architecture, Cassandra, Ansible, Bash Script, DNS Servers, Infrastructure Architecture, IPsec, Google BigQuery, Google Cloud Dataproc, Spark, AWS Cloud Architecture, Agile, Digital Transformation, AWS Cloud Computing Services, Monitoring, Cloudflare, Amazon RDS, Ethereum, Orchestration, Grafana, Containerization, Git

Infrastructure Architect | DevOps Team Lead

2013 - 2016
Iponweb.com
  • Designed the infrastructure in the cloud and wrote own provisioner.
  • Planned a company-wide technology migration, including big data pipelines.
  • Investigated and compared cloud technology Google Cloud vs. Amazon.
  • Invented and implemented an in-house load-balancing solution.
  • Implemented and invented a company-wide build software policy.
  • Integrated an Open Build Service into the company's delivery processes.
  • Designed a global DNS with the main goal of minimizing resolution times around the world.
  • Prepared an in-house CDN for internal use, such as package delivery to thousands of servers.
  • Re-worked and re-designed the approach for using the Puppet automation framework.
Technologies: Amazon Web Services (AWS), Networking, Puppet, Google Cloud Platform (GCP), SQL, MariaDB, MySQL, System Administration, Serverless, SQLBase, PostgreSQL, DevOps, VPN, OpenStack, CI/CD Pipelines, Redis, Java, Prometheus, Apache Kafka, NGINX, DNS, Emailers, NAT, Infrastructure as Code (IaC), Docker, APIs, Cloud Architecture, Amazon S3 (AWS S3), LAMP, Rational Software Architecture, Cassandra, Ansible, Bash Script, DNS Servers, Infrastructure Architecture, IPsec, Digital Advertising, Agile, Linux Administration, Web Security, AWS Cloud Computing Services, Team Management, Monitoring, AWS Auto Scaling, Orchestration, Grafana, Containerization, Git, IntelliJ IDEA

Lead Systems Operations

2011 - 2013
sputnik.ru
  • Managed a team of five on the systems operations team.
  • Designed the backbone network for the Hadoop cluster.
  • Created an automation solution, including style guide, tags policy, and pre- and post-commit hooks for Puppet.
  • Designed a global DNS solution, including Geo DNS balancing.
  • Investigated performance issues in the network, including bonding MPLS.
  • Designed and implemented different layers of redundancy for the search engine application.
Technologies: Puppet, Hadoop, SQL, MariaDB, MySQL, System Administration, Serverless, SQLBase, PostgreSQL, DevOps, VPN, OpenStack, Prometheus, NGINX, Email, Emailers, NAT, Infrastructure as Code (IaC), APIs, LAMP, Cassandra, Ansible, Bash Script, DNS Servers, Infrastructure Architecture, IPsec, Agile, Linux Administration, Monitoring, Orchestration, Grafana, Containerization, Git, IntelliJ IDEA

Senior Systems Administrator

2010 - 2011
imhonet.ru
  • Managed a 3-person ops team, built safe dev/release environments, designed front/backbone networks for a high-load web app, created an internal libvirt+KVM cloud, and rolled out Zabbix-based monitoring and alerting.
  • Introduced a new monitoring and alerting approach with Zabbix.
  • Described the production infrastructure with Puppet and stored it in Git.
Technologies: Zabbix, KVM, libvirt, Puppet, PHP, Networking, Scripting, Perl, Bash, Linux, SQL, MariaDB, MySQL, System Administration, Serverless, SQLBase, DevOps, VPN, NGINX, DNS, Email, Emailers, NAT, APIs, LAMP, Ansible, Bash Script, DNS Servers, IPsec, Monitoring

Automation Engineer | Lead Sysop

2009 - 2010
Chronopay
  • Invented and implemented shared hosting for partners.
  • Automated a routine with Python, Bash, Perl, and scripts.
  • Automated the deployment process with a different set of in-house scripts.
  • Prepared for security scans, pen tests, and audit reporting.
  • Migrated infrastructure to new Linux/Ubuntu distribution.
Technologies: Ubuntu, Python, Networking, Scripting, Perl, Bash, Linux, SQL, MariaDB, MySQL, System Administration, Serverless, VPN, NGINX, DNS, Emailers, NAT, APIs, LAMP, Bash Script, DNS Servers, IPsec, Monitoring

Systems Operations Engineer

2007 - 2009
Syntag
  • Integrated the system for different security solutions based on PCI DSS.
  • Prepared PCI DSS for our clients. and other compliances.
  • Completed penetration testing for different clients.
  • Set up and managed LDAP servers for Linux-based servers.
  • Designed and implemented mail servers with Postfix, Exim, Dovecot, and CirrusDB.
Technologies: Exim4, Web Servers, Networking, Proxy Servers, SQL, MariaDB, MySQL, System Administration, Serverless, VPN, NGINX, Emailers, NAT, Bash Script, DNS Servers, IPsec, Monitoring

Experience

Proprietary Trading Company

I set up a hybrid infrastructure across AWS and bare-metal data centers to train machine learning models and deliver them to trading systems with ultra-low latency. Built data pipelines to deliver order book data for model training and to rapidly update trading models across AWS and on-premises broker colocations.

Public Talk

https://www.youtube.com/watch?v=eflA1mUdmiA
A talk about my load balancing solution (only available in Russian). Multiple layers of complex balancing and problems were faced during the implementation.

Takeaway: how to be reliable and predictable under the heavy load (11 million requests/minute).

Skin Assistant

https://geltek.ru/ai-skin/
End-to-end delivery and management of a YOLO-based, self-trained AI model for skin anomaly detection — from concept, data preparation, and training to process setup and hosted inference — including full development, integration work, API architecture, and cloud deployment. Unfortunately, the site is in Russian. At this time, we have around 3-2 inferences per second.

Education

2000 - 2006

Master of Science Degree in Organic Chemistry

Academy of Fine Chemistry - Moscow, Russia

Skills

Libraries/APIs

libvirt

Tools

Git, IntelliJ IDEA, Zabbix, Helm, Puppet, Ansible, Terraform, AWS CloudFormation, ELK (Elastic Stack), Grafana, VPN, Google Cloud Dataproc, Google Kubernetes Engine (GKE), GitHub, NGINX, Google Workspace, Amazon EKS, GCP Network, Amazon SageMaker, Jupyter, You Only Look Once (YOLO), Amazon Elastic Container Service (ECS), Amazon Elastic Container Registry (ECR), Exim4, CVAT

Languages

Bash, Python, Bash Script, Java, SQL, Perl, PHP, Erlang, Go, Rust, Solidity, Lua, C

Frameworks

Spark, AWS HA, Hadoop

Paradigms

Continuous Integration (CI), DevOps, Agile, ETL, Rational Software Architecture, Continuous Delivery (CD), Microservices Architecture

Platforms

Ubuntu, Google Cloud Engine, Amazon Web Services (AWS), Google Cloud Platform (GCP), Linux, Kubernetes, OpenStack, Drone CI, Visual Studio Code (VS Code), Docker, Apache Kafka, AWS Cloud Computing Services, LAMP, AWS Lambda, Firebase, Ethereum, KVM, PagerDuty

Storage

Amazon DynamoDB, Google Cloud, BigTable, Elasticsearch, Google Cloud SQL, SQLBase, Redshift, MongoDB, MySQL, PostgreSQL, Cassandra, CockroachDB, MariaDB, Redis, Amazon S3 (AWS S3), Datadog

Other

Proxy Servers, Networking, Web Servers, Scripting, Cloud, Migration, Networks, DNS Servers, Load Balancers, Infrastructure Architecture, TICK Stack, IPsec, Digital Advertising, Google BigQuery, EMR, Serverless, Interconnected Networks, Team Management, Digital Transformation, AWS Cloud Architecture, AWS Auto Scaling, System Administration, Pulumi, CI/CD Pipelines, Monitoring, Infrastructure as Code (IaC), Linux Administration, Containerization, Orchestration, Web Security, Cloudflare, Prometheus, Emailers, DNS, Email, Office 365, NAT, Analytics, API Integration, Machine Learning Operations (MLOps), APIs, Cloud Architecture, AlloyDB, Cloud Migration, FastAPI, Argo CD, GitOps, GPU Computing, Serverless GPUs, gpu operator, Amazon SageMaker Pipelines, Data Inference, AI Learning, GitHub Actions, SOC 2, Amazon RDS, Organic Chemistry, Team Leadership, Leadership, Linux HPC, Consensus Algorithms, Cloud Computing, IT Projects, Startups, Full-stack, Big Data

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring