SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
inDrive operates ML models across 48 countries with a rapidly growing platform. The company deploys to Databricks, SageMaker, and self-managed GPU inference, with LLM workloads moving into production.
You will join as a Senior Backend Engineer building and operating the services and tooling that ML teams depend on. You will own services end-to-end from design through production on-call, writing production-grade Go and Python. You'll learn the MLOps stack—Terraform, Kubernetes, Databricks, SageMaker—on the job while collaborating with data scientists, ML engineers, and platform engineers.
Key responsibilities:
- Design, build, and run backend services and internal tooling for the ML platform in Go and Python, owning delivery, monitoring, and on-call support.
- Drive technical research and design for new platform features, documenting trade-offs.
- Build and extend deployment tooling and SDKs for ML teams to ship models, including serving, autoscaling, and data capture.
- Integrate ML services with the wider backend using sync (HTTP, gRPC) and async (Kafka, RabbitMQ) communication, optimized for reliability under high load.
- Support the full ML delivery cycle: CI/CD for models and pipelines, monitoring, and model performance metrics.
- Build observability end-to-end: metrics, logs, dashboards, tracing, alerting, and runbooks.
- Write infrastructure as code with Terraform and Terragrunt, shipping through reviewed pull requests.
- Participate in code and design reviews, mentor engineers, and maintain review standards for AI-generated changes.
Requirements:
- 5+ years of commercial backend engineering, including high-load systems.
- Strong Go with production experience using advanced features.
- Solid Python for data processing (batch, streaming, async) and service code.
- Strong SQL: schema design, indexes, query optimization, replication, and partitioning. Experience with Redis or similar key-value stores.
- Experience with sync and async service communication (HTTP, gRPC, Kafka, RabbitMQ) with focus on reliability.
- Docker, CI/CD pipelines, and working Kubernetes knowledge. AWS/GCP core services (compute, storage, networking, IAM).
- Understanding of ML lifecycle: data preparation, model performance metrics, non-functional requirements, concept drift. Familiarity with scikit-learn, PyTorch/TensorFlow.
- Experience designing and running observability: metrics, logging, dashboards, tracing (Prometheus, Grafana, ELK or similar).
- Git workflows, code review, Linux command line, and networking basics (TCP/IP, DNS, VPC, load balancers).
- Ability to explain trade-offs clearly to data scientists, product owners, and on-call engineers.
Nice to have:
- Hands-on MLOps tooling: Databricks (Unity Catalog, Asset Bundles), SageMaker, Kubeflow, MLflow, Airflow.
- GPU or LLM inference experience (vLLM, Triton, Langfuse).
- Experience using AI coding agents with full review of generated changes.
- Experience in emerging markets or building without established playbooks.
- Infrastructure-as-Code (IaC) tools experience.
