SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Twelve Labs is building AI infrastructure to enable machines to understand video at scale. Video comprises 90% of global data, yet most remains inaccessible to machines. The company has raised over $300M from leading investors including NEA, Radical Ventures, Amazon, NVIDIA, Snowflake, and others, with advisors including Fei-Fei Li, Silvio Savarese, and Alexandr Wang. Twelve Labs operates globally from San Francisco, Seoul, New York, and London, with core R&D in Seoul.
The Construction team (Video Ingestion & Serving Platform) develops end-to-end infrastructure for processing customer video from upload through AI model consumption. The platform handles video collection, decoding, embedding, metadata processing, storage, and model serving across large-scale multitenant SaaS and enterprise deployments. The team uses Python and Go services, Temporal, PostgreSQL, Kubernetes, KServe/vLLM, Terraform, ArgoCD, and Grafana/OpenTelemetry.
This Staff ML Engineer role will lead the technical direction and production operations of the Video Ingestion & Serving Platform. You will design, develop, and operate backend services and platforms for large-scale video processing, embedding, and AI model serving. Responsibilities include improving concurrent execution, retry logic, backpressure, and failure recovery for long-running Temporal-based workflows; optimizing throughput, latency, GPU utilization, and infrastructure costs through load testing, profiling, and operational metrics; designing PostgreSQL data models, sharding strategies, and query paths while leading zero-downtime production operations; building and improving Kubernetes, Terraform, and ArgoCD-based infrastructure and CI/CD pipelines using Karpenter and KEDA for scalability; advancing observability (metrics, traces, logs, alerting) and incident response to improve service reliability; and collaborating with backend, ML, infrastructure, and product teams to set technical direction and lead production rollouts.
As a Staff Engineer, you will write code directly, resolve production issues, and create technical standards and platform capabilities for cross-team use. You are expected to trace problems to their root cause—whether in services, databases, workflows, or Kubernetes—and build sustainable solutions.
REQUIREMENTS:
- 5+ years of software engineering experience or equivalent capability
- Experience designing and operating distributed systems in production, including workflows, queues, and databases
- Proficiency in Go or Python with practical development capability in other languages
- Direct experience operating services on Kubernetes and cloud infrastructure with Infrastructure-as-Code automation (e.g., Terraform)
- Experience performing live migrations of databases, storage, or backend systems while considering downtime and data consistency
- Experience identifying performance bottlenecks through load testing, profiling, and monitoring, with demonstrated improvements in throughput or cost
- Experience leading technical decisions across multiple teams, conducting design reviews, and mentoring to increase team execution
- Ability to take high ownership in fast-changing environments, driving problem definition through production operations
PREFERRED QUALIFICATIONS:
- Experience building or operating GPU-based inference serving systems (KServe, vLLM, Triton)
- Experience with durable workflow engines (Temporal, Cadence, Step Functions)
- Experience with sharded or distributed databases (Aurora Limitless, Citus, Vitess)
- Experience building and operating observability stacks (Grafana, Mimir, Loki, Alloy, OpenTelemetry)
- Experience with FFmpeg, video decoding, transcoding, or large-scale media processing pipelines
- Strong English communication skills for global team collaboration
