King's Choice

We're looking for an SRE / DevOps engineer to build the reliability layer for a platform that commands live power plants. No unified log pipeline, no dashboard anyone watches, no rehearsed restore - the app runs in production, the layer underneath it is a set of habits. You decide what the observability stack looks like, what an alert may wake a person up for, and how failover is designed and tested.

Second part is physical: with the Infra Lead, dispatch control room - workstations, UPS, local network, secured remote access to plants. Grafana in one tab, a rack in the other.

Must have

• 4+ years SRE / DevOps / systems engineering, Linux daily: services, systemd, storage, performance diagnostics
• Real experience with physical or VPS servers, not cloud consoles only - UPS, RAID, remote management. Cloud-only profile is an automatic no
• Cloud VMs: provisioning, patching, hardening, capacity planning, TLS, nginx
• Docker / Docker Compose in production: dev/staging/prod layouts, registry, zero-downtime deploys
• Observability you built yourself: Prometheus + Grafana or equivalent, log pipeline (fluent-bit / ELK class), and an alerting philosophy you can defend
• Redundancy you tested: PostgreSQL backup/restore and replication, Redis failover, measured recovery. You can describe a backup you personally restored
• Networking: WireGuard or similar, firewalls, DNS, TLS, diagnosing cloud ↔ office ↔ plant links
• CI/CD as code (GitHub Actions or similar), promotion through environments
• Bash + Python for automation
• English B2

We're looking for a Lead Python Engineer who will be the key developer on the team - someone who has already designed architecture, not just built against someone else's.

If an e-commerce site goes down, it's inconvenient. This product has to run without failure: if a software glitch shuts down a 50 MW plant, that means direct financial losses and safety risks. Your code moves megawatts, not pixels or shopping carts.

About the company

A European energy company that builds and operates power plants: solar, wind, and industrial-scale battery energy storage systems.

• Our own EMS platform started as an internal tool and is now being turned into a standalone product for any plant owner or energy service company
• The product is at MVP stage

What you'll do

• Take over the platform, together with a small team (a Md Django engineer is already on board)
• Run a technical audit of the MVP
• Design and implement failover for plant connections: reconnect, buffering, redundancy - so telemetry and commands survive unreliable 4G links
• Build agentic pipelines to automate development, testing and deployment, and make Spec-Driven Development a mandatory standard
• Work in two modes: Rapid Validation (shipping lean features to test hypotheses) and Stabilization (hardening what shows traction)
• Do technical pruning: remove dead code and features behind hypotheses that didn't work out
• Set engineering standards: code review, branching, CI gates (ruff, pytest), definition of done. The API has to stay stable for a dispatch center that runs 24/7

Who we're looking for

• 8+ years of production Python experience
• The full stack in production: Django + DRF, FastAPI (Pydantic v2), Celery, PostgreSQL, Redis, Docker / Docker Compose, pytest + CI
• API and system architecture design: service boundaries, contracts, data flows - not only implementing against someone else's design
• RBAC: role models on the principle of least privilege, DRF permissions, object-level access, access enforced consistently across services
• Integrating external APIs behind a caching layer (Redis): cache strategy and invalidation, rate limits, retries/backoff, serving stale data gracefully when the upstream is down
• Failover thinking: reconnect strategies, buffering, health checks, redundancy, graceful degradation over unreliable links
• Spec-Driven Development: contract-first (OpenAPI/Swagger, Pydantic, gRPC), specifications as the source of truth for AI agents
• Agentic pipelines: CrewAI, LangGraph or custom orchestrators, automating engineering workflows
• English B2+

Soft skills

• Product mindset. We need someone who treats the platform as their own product, not as another outsourcing project. Code is a temporary asset until the business model is validated
• Ownership. The result is production, not a PR
• Ability to explain an architectural decision to a non-engineer, and to stay calm during incidents