Backend engineer

About the job

You’ll work at the core of our production platform, building and operating backend services that power APIs, product features, and ML inference systems.

This is a mid-level backend role focused on reliable service delivery, API quality, and production ML-serving infrastructure. You’ll primarily build in Go, contribute occasionally in Rust or Python, and support cloud-native systems running mostly on AWS (with some GCP).

You won’t be doing ML research — but you will be responsible for making trained models work reliably in real products.

Your key responsibilities

• Build and evolve production backend systems
  • Design, implement, and maintain backend services that power customer-facing and internal capabilities.
• Develop high-quality APIs
  • Create and maintain APIs with strong contracts, documentation, versioning, and backward compatibility practices.
• Productionize ML inference systems (non-research)
  • Integrate trained model artifacts into backend services and release workflows. Build and operate inference endpoints and supporting systems such as model routing, versioning, caching, and feature retrieval.
• Improve reliability and performance
  • Monitor and optimize latency, throughput, error rates, and infrastructure cost across backend and inference services.
• Maintain engineering quality
  • Write appropriate unit and integration tests, participate in code reviews, and follow team standards for documentation and style. Debug issues across environments and contribute to root-cause analysis and prevention.
• Operate services in the cloud
  • Deploy, monitor, and scale services primarily in AWS (and some GCP). Contribute to CI/CD workflows and release safety improvements.
• Support operational readiness
  • Add observability (metrics, logs, traces), assist with incident response, and follow secure engineering practices including proper auth patterns and secrets handling.
• Collaborate across teams
  • Work closely with ML and data engineers on serving requirements, data contracts, and operational quality — focused on inference and serving, not training.

What makes this role different

• You won’t just build CRUD services — you’ll help run ML-powered systems in production

• You’ll work directly on inference infrastructure: model serving, routing, performance, and cost optimization

• You’ll contribute to both product APIs and intelligence-serving systems

• You’ll have meaningful ownership of production services and their reliability

• Startup mindset required — be ready to adapt, take ownership, and occasionally contribute outside your core lane (for example small SDK or dashboard support in TypeScript when needed)

This role is ideal for someone who

• Enjoys building production backend systems that are actually used by customers

• Cares about API quality, reliability, and operational excellence

• Likes performance tuning and debugging real systems

• Is interested in ML-powered products and inference systems, even without doing ML research

• Balances delivery speed with maintainability and clean design

• Works well in collaborative design and code review environments

Required qualifications

• 2–5 years (or equivalent) professional experience building backend systems in production

• Strong hands-on experience with Go, including concurrency, debugging, profiling, and performance-aware development

• Experience building and maintaining APIs consumed by other teams or customers

• Experience with relational and/or NoSQL databases and common data-access patterns

• Experience operating services in cloud environments, ideally AWS (compute, networking basics, IAM, monitoring)

• Practical familiarity with ML inference concepts such as model versioning, latency/throughput tradeoffs, and monitoring (no research background required)

• Hands-on experience operating ML inference services (for example PyTorch or ONNX), ideally in recommendation, semantic search, or personalization systems

• Strong engineering practices: testing, code review, documentation, and collaborative delivery

Preferred qualifications

• Exposure to Rust and/or Python in production settings

• Familiarity with distributed systems patterns (queues, streams, retries, idempotency, eventual consistency)

• Experience with containers, Kubernetes, and/or infrastructure-as-code (Terraform or CDK)

• Experience with ecommerce or event-driven product systems (catalog, personalization, analytics, experimentation)

• Familiarity with security and privacy practices in multi-tenant SaaS systems

• Experience contributing to SDKs or developer-facing tooling