Full-Stack Engineer — Temporal AI Platform

Role overview Build an end-to-end platform that makes large temporal foundation models usable in production. The work spans backend services, data and inference infrastructure, and cloud operations, with close collaboration with researchers to turn experimental model capabilities into dependable forecasting workflows.

Responsibilities - Build platform services for data ingestion and management, model inference, forecasting, and delivery of results. - Connect models with data pipelines, serving infrastructure, APIs, and user-facing applications. - Develop reliable, low-latency inference services that support different datasets, models, forecast horizons, and users. - Create systems to ingest, store, query, transform, and process large time-series, tabular, and structured datasets. - Deploy and operate cloud services and machine-learning workloads, monitoring reliability, performance, scale, and cost. - Work with researchers and ML engineers to productionize models and shape complete user workflows.

Requirements - Strong full-stack engineering experience across backend systems, APIs, data infrastructure, and cloud services. - Ability to build scalable, reliable services and reason about latency, throughput, observability, and operational needs. - Practical understanding of model inputs and outputs, inference workflows, and common model failure modes. - Experience turning experimental ML models or research code into robust production services in collaboration with ML teams. - Product judgment and comfort adapting as research, infrastructure, and user needs evolve.

Nice to have - Experience with ML platforms, data or analytics products, developer tools, or AI applications. - Background serving foundation or deep-learning models; familiarity with GPUs, batching, job queues, caching, or distributed inference. - Cloud operations experience with tools such as Docker, Kubernetes, Terraform, CI/CD, and observability platforms. - Experience with structured or time-series data, PyTorch, forecasting, experimentation, or model evaluation. - Familiarity with LLM or agent workflows, including retrieval, tool use, streaming, or long-running asynchronous tasks.

Benefits and work setup Full-time, remote role open to candidates in the US or globally.