BitDeer Technologies Group

Join Bitdeer’s SRE/Monitoring Platform team as an entry-level software engineer building the observability and automation substrate for a global GPU cloud. You’ll develop production components for telemetry collection, alerting, topology, remediation, and SLOs while learning under senior engineers.

Launch your AI engineering career by building large-scale cloud, Kubernetes, model-serving, and AI agent platforms at Bitdeer AI. New graduates can join areas ranging from cloud infrastructure and MLOps to AI applications, security, and research, with opportunities to rotate across roles.

Own the measurement and modeling systems that make Bitdeer AI’s MaaS inference benchmarks, capacity plans, latency metrics, and cost projections reflect real production workloads. This role combines production traffic analysis, LLM serving expertise, benchmarking fidelity, and cross-functional performance reporting.

Build and operate the integration layer connecting Bitdeer AI’s MaaS platform with third-party inference providers. You’ll normalize provider APIs, implement cost controls and circuit breakers, monitor contract changes, and reconcile usage with vendor invoices.

Own the developer-facing infrastructure-as-code experience for Bitdeer’s GPU cloud, building and evolving its Terraform provider, CLI, and SDKs. You’ll shape the product surface, compatibility strategy, and release lifecycle while partnering closely with the evolving Kubernetes control-plane API.

Research and build next-generation LLM inference systems that deliver measurable improvements in production serving fleets. This role combines frontier systems research with hands-on implementation, with results expected to ship and potentially publish.

Research Engineer internship focused on making AI models faster and cheaper to serve without sacrificing meaningful quality. You will implement and develop model-efficiency techniques, rigorously evaluate results, and own a project producing a real outcome within 3–6 months.

Research Engineer focused on making large language model inference faster and more cost-efficient without sacrificing meaningful quality. You will implement and develop model-efficiency techniques across models and hardware, then rigorously evaluate their production impact.