Distributed System Engineer

About SimForge SimForge is building the data infrastructure for Physical AI. We create photorealistic digital twins, HD semantic maps, and cloud-native simulation infrastructure that enable autonomous vehicles, robotics, and smart city AI systems to train, test, and validate safely at scale. Headquartered in San Francisco with offices in Luxembourg and Goa, SimForge is led by autonomous vehicle industry veterans from Amazon, Cruise, Uber, and Mapbox. We are looking for a distributed systems engineer to own the orchestration layer that makes this run reliably across cloud GPU providers.

What you’ll own You will design and build the core job orchestration system for SimCloud: the layer that turns simulation requests into reliable, observable, cost-aware, reproducible workloads across multiple GPU-backed cloud₹ environments. This includes job scheduling, provider abstraction, worker lifecycle management, queueing, retries, artifact movement, quota management, execution state tracking, and integrations with simulation runtimes. You should be comfortable operating at the boundary of backend systems, cloud infrastructure, GPU workloads, and developer tooling.

Responsibilities Design and build distributed orchestration systems for large-scale simulation jobs across multiple cloud GPU providers. Create abstractions for launching, monitoring, retrying, cancelling, and resuming simulation workloads. Build provider-agnostic interfaces over cloud GPU capacity, storage, networking, identity, and execution environments. Own reliability primitives: idempotency, retry semantics, durable state, failure isolation, backpressure, and graceful degradation. Build observability for simulation runs, including logs, metrics, traces, cost visibility, artifact status, and user-facing execution state. Design data and artifact pipelines for moving simulation outputs into datasets, manifests, and downstream evaluation workflows. Improve execution efficiency through batching, caching, spot/preemptible capacity, quota-aware scheduling, and cost/performance tradeoffs. Partner with simulation, ML, and product engineers to make SimCloud feel fast, reliable, and debuggable. Use AI coding tools and agentic workflows as part of your day-to-day engineering loop, including code generation, test harnesses, debugging agents, infra audits, and reliability experiments. Help define engineering standards for distributed execution, reproducibility, security, and operational excellence.

What we’re looking for Strong experience building distributed systems, job orchestration platforms, workflow engines, compute platforms, data platforms, or cloud infrastructure. Deep backend engineering ability in one or more of Go, Python, Rust, Java, or TypeScript. Experience with Kubernetes, containers, queues, workflow engines, cloud compute, object storage, and infrastructure-as-code. Comfort debugging production systems with complex failure modes. Experience designing APIs and internal platforms used by other engineers. Strong opinions about reliability, observability, reproducibility, and operational simplicity. Ability to move quickly in ambiguous product and infrastructure spaces. Hands-on fluency with AI-native development workflows: coding agents, test generation, evaluation loops, promptable dev tools, CI automation, and agent harnesses.

Nice to Have Experience with GPU infrastructure, simulation platforms, robotics, AV simulation, game engines, rendering farms, ML training infrastructure, or synthetic data pipelines. Experience across multiple cloud providers or GPU providers. Experience with Ray, Temporal, Argo, Airflow, Nomad, Slurm, Kubernetes operators, or custom schedulers. Experience with large artifact pipelines, dataset generation, simulation determinism, or reproducible compute. Experience building agentic developer workflows or internal AI tools for engineering teams. Location - Goa, India Salary - Rs 18,00,000 to Rs 35,00,000 CTC + ESOPs