Join
the team

What you will do

• Build the team and set up the process: from hypothesis to a model in production
• Set the roadmap: anti-fraud on payments, anomalies in traffic, automatic node selection for users
• Answer to the owner for the function's metrics, not for the number of experiments
• Choose the training and deployment infrastructure and keep it from turning into a zoo

What we expect

• Two years or more leading an ML team, with a strong engineering background of your own
• Production ML systems you took all the way into operation, not ones that stopped on a laptop
• Python and the classic stack (scikit-learn, XGBoost/LightGBM), a grasp of binary classification metrics
• The ability to explain to the business why a model is wrong, and what that costs in money

Nice to have

• Anti-fraud or payment scoring
• Streaming data: Kafka, windowed aggregation, real-time inference
• MLOps: Docker, Kubernetes, Airflow, MLflow
• Experience hiring and mentoring

What you will do

• Build anomaly detection models on connection and payment logs
• Build automatic server selection: predict where a person will be faster
• Take a model all the way to production and watch it degrade, rather than filing it away
• Prepare the data: marts, features, labelling

What we expect

• Python and the ML stack: scikit-learn, XGBoost/LightGBM; PyTorch where the task calls for it
• Confident SQL and experience with large volumes of unstructured data: raw logs, event streams
• A grasp of quality metrics and the ability to pick a threshold against the business cost of an error
• Git, code review, reproducible experiments

Nice to have

• Experience with network or payment data
• Airflow and pipeline orchestration
• Docker and shipping models as a service
• Publications, Kaggle, open source

What you will do

• Deploy and maintain nodes: configuration, updates, automating the routine
• Run CI/CD: build, deploy, roll back with no manual steps and no heroics at night
• Set up observability — metrics, logs, alerts, incident on-call
• Count what the infrastructure costs and bring it down without losing headroom

What we expect

• Linux at the level of “I will work the incident out myself”, networks at the level of routes and NAT
• Kubernetes and containers in production, Terraform or OpenTofu, Ansible
• Prometheus and Grafana (or VictoriaMetrics), sensible alerts instead of a wall of graphs
• One language for automation: Go or Python

Nice to have

• Experience with tunnelling protocols and network services under load
• Several clouds and bare metal at the same time
• Security: secrets, access, audit
• Experience with on-call rotas and postmortems

What you will do

• Configure and maintain nodes: routing, balancing, fault tolerance
• Work out degradations: where packets were lost and why latency grew
• Work with the sites: connections, links, expansion plans
• Measure and document — so that the promises on the site are backed by a measurement

What we expect

• Networks in earnest: TCP/IP, BGP, routing, NAT, MTU and everything that breaks in practice
• WireGuard and tunnelling protocols, iptables/nftables
• Linux, configuration automation, tcpdump and reading traffic
• Willingness to take on-call shifts along with the team

Nice to have

• Experience running a network across several data centres and countries
• Anycast, L4/L7 load balancers
• Experience with high-load proxy services
• Scripts and automation in Python or Go

What you will do

• Answer on Telegram and by email: work through connections, payments, devices
• Reproduce the problem and hand engineers a clear report, not “it does not work for a customer”
• Write and maintain guides, so the next question like it never arrives
• Notice repeat requests and bring them to the product as a task

What we expect

• Written Russian that is calm and human — no officialese
• An everyday understanding of how the internet and networks work: IP, DNS, router, Wi-Fi
• Experience supporting users, or a strong wish to get into it
• Willingness to work shifts, including weekends on a rota

Nice to have

• Experience with tunnelling clients and configuring network applications
• English for correspondence with vendors
• Experience in SaaS or hosting support
• The knack of writing instructions people actually follow

What you will do

• Run hiring from request to offer: sourcing, screening, coordinating interviews
• Own onboarding: the first day, the first month, feedback
• Grow the employer brand: job texts, channels, presence in the right communities
• Measure the hiring funnel and talk about it in numbers

What we expect

• Experience hiring in IT and an understanding of the difference between DevOps, ML and support
• The ability to write job posts people read, and rejections that leave the relationship intact
• Self-direction: the team is distributed and nobody stands over you
• Care with data: ATS, deadlines, candidate statuses

Nice to have

• Experience in a distributed team
• Knowledge of the remote market and its grades
• English for working with candidates outside Russia
• Experience building HR processes from scratch