Systems Engineer

About TubeScience

TubeScience is one of the fastest-growing performance video companies in the world. We use data, AI, and creative systems to produce video advertising at a scale and speed that no one else has figured out.

Our engineering team builds the infrastructure that makes this possible — from AI agent pipelines to the ad operations tooling that drives hundreds of millions in client spend.

We are profitable, growing fast, and building in a space where the technical problems are genuinely challenging. You will be designing and owning systems that directly move business outcomes.

The Role

We're hiring a Systems Engineer to own the infrastructure that makes our AI platform reliable in production. That means the media pipelines, LLM routing layers, distributed data systems, and observability tooling that everything else depends on.

This isn't a role where you execute someone else's architecture. You'll be one of a small number of engineers who decides how these systems are built — and then builds them. The work ships fast, the feedback loop is short, and the impact is direct.

This position is fully remote, with meaningful overlap during afternoon hours in your timezone (CET/EET).

What You'll Own

• Media ingestion platform — taking raw uploads all the way through validation, transcoding, staging, and lifecycle-aware storage management

• LLM infrastructure layer — routing requests across providers, normalizing outputs, applying semantic caching, and keeping things reliable under load

• Data platform — ingesting and indexing multi-modal data, powering retrieval and search, managing context assembly, and maintaining long-term data health

• Agent and platform operations layer — MCP servers, A2A dispatch, permissions infrastructure, and the observability layer that keeps everything visible and accountable

Who You Are

• You instrument and observe distributed systems end to end — not just reading dashboards, but designing for observability from the start

• You've built and operated data engineering and ETL pipelines in production

• You design clean, durable APIs across REST, gRPC, event-driven, and schema-based systems — comfortable with MCP, A2A, schema registries, and tool-calling normalization

• You know media pipelines: transcoding, ffmpeg-like tooling, cloud storage, and large-object processing

• Solid hands-on experience with PostgreSQL, MongoDB, Redis, and cloud storage-backed systems

• Comfortable with Docker, Kubernetes, Terraform, or Pulumi

Bonus Points

• Experience integrating LLM/AI systems in production: working with LLM APIs, provider routing, multi-provider abstraction, schema normalization across providers, structured output, semantic caching, agentic AI systems, and prompt engineering at the infrastructure layer