Senior DevOps Engineer at Vse AI

⭐️ AI startup ⭐️ Full remote or office in Saint Petersburg ⭐️ Potential future growth to Head of Infrastructure

Vse AI is an aggregator of the world's leading neural networks, available directly in Telegram. In December, we successfully conducted a beta launch and are now in an active growth phase. The product is already live, with real users and load, and our immediate goal is to confidently handle multiple growth (10x) without service quality degradation.

We are looking for a Senior DevOps Engineer who will take responsibility for the infrastructure, stabilize it under peak loads, and essentially design a new, more resilient foundation for further product scaling.

🔶 You're a good fit for us if: • you have 4+ years of commercial experience in DevOps; • you have solid production experience with GCP, including Cloud Run; • you understand how Python / FastAPI applications work in production (Uvicorn, Gunicorn, specifics of deployment and scaling); • you are comfortable entering a system where: – not all processes are formalized yet, – there are real problems, and engineering solutions are expected from you, – you can work autonomously, make architectural decisions, and justify them.

🔶 Responsibilities: 1️⃣Redesign and stabilize the infrastructure on Cloud Run: • resolve auto-scaling issues, • achieve predictable behavior under sharp and irregular load spikes, • minimize outages and instability. 2️⃣Build infrastructure practically from scratch: • propose and implement an infrastructure-as-code approach (Terraform / Pulumi / other tools — your choice), • bring order to configurations and environments. 3️⃣Design and implement CI/CD: • currently, the deployment process is not documented and was done manually, • we expect you to build a transparent and reproducible pipeline based on the product's real needs. 4️⃣Set up monitoring and observability: • logs, metrics, alerts, • the ability to quickly understand what exactly is breaking and why. 5️⃣Optimize deployment and operation of FastAPI applications considering Cloud Run specifics and external AI providers. 6️⃣Work with unstable scenarios: • inconsistent responses from external APIs, • issues with model versions, • errors that only appear under load or during scaling.

🔶 Conditions and format: 1️⃣Remote work without geographical restrictions or office in Saint Petersburg. 2️⃣Project-based cooperation format: • you influence the cost of your services yourself, • payment is made in two stages: at the start and upon completion of the agreed scope of work. 3️⃣ Small team, direct communication with founders, quick decisions without bureaucracy. 4️⃣ Upon successful completion of tasks and product growth, a transition to a permanent format and further development of the role up to Head of Infrastructure is possible.