Top Tech Jobs & Startup Jobs

Designs, deploys, and operates high-performance InfiniBand and RoCE networking fabrics for GPU clusters, customer connectivity, and multi-site WAN infrastructure. Manages routing, switching, BGP peering, VLAN/VRF/VPC architectures, automation, monitoring, troubleshooting, documentation, capacity planning, and security audits. Requires deep data center or service provider networking expertise, advanced routing knowledge, and hands-on experience with Arista, NVIDIA, or Cisco platforms.

Lead the reliability and scaling of Horizon AI’s autonomous platform for production GPU infrastructure. Harden agent dispatch with circuit breakers, retries, health-aware routing, and observable error handling. Build caching and event-streaming capabilities, determine and implement the orchestration strategy, introduce infrastructure as code, enable concurrent agent sessions, and own service standards, testing, CI, observability, and deployment practices. Serve as technical deputy and collaborate with Systems and SRE teams.

Build and scale Horizon AI’s platform reliability through caching, event streaming, resilient agent dispatch, observability, infrastructure as code, and safer deployment practices. The role focuses on health checks, circuit breakers, retries, backpressure, tracing, alerting, canary releases, traffic shifting, and rollback strategies. It requires measurable experience improving distributed-system reliability in Python and containerized environments, with Kubernetes or comparable orchestration expertise.

Lead development of a unified fleet knowledge model and decision-capture pipeline for autonomous data-center operations. Build data access layers across relational, telemetry, monitoring, and time-series systems; capture labeled operator decisions; negotiate data requirements with source-system owners; and own scaling considerations including vector indexing, schema evolution, retention, and analytics separation. The role requires strong infrastructure-domain modeling, PostgreSQL expertise, and experience handling time-series and event data at scale.

Leads GPU performance optimization for customer training and inference workloads. Establishes hardware-specific performance baselines, identifies bottlenecks, tunes distributed workloads, and converts optimization expertise into agent-executable improvements. Partners with customers on workload tuning, while secondarily owning internal model serving, cost- and latency-aware routing, model migration, and GPU allocation decisions.

Manages the operational, technical, and lifecycle success of assigned managed-services customers. Responsibilities include customer relationship management, service reviews, ticket and escalation oversight, asset lifecycle planning, renewals, contract and licensing coordination, initiative tracking, documentation, account health reporting, and sales opportunity handoffs. The role coordinates across service delivery, engineering, PMO, sales, procurement, finance, and leadership while communicating with customer executives and technical stakeholders.

Build and operate the multi-tenant platform layer for a GPU cloud service. Responsibilities include Kubernetes-based orchestration, tenant isolation, platform APIs and tools, GPU cluster automation, infrastructure as code, reliability and observability partnerships, customer issue support, environment templates, rollout tooling, architecture reviews, and reducing operational toil.

Owns reliability, observability, and incident response for a GPUaaS platform. Defines SLOs, builds monitoring and alerting systems, leads major incidents and post-incident reviews, automates operational processes, maintains runbooks, manages on-call operations, coordinates with engineering teams, drives chaos testing, reports SLA performance, and mentors junior engineers.

Monitor GPUaaS infrastructure, triage alerts, execute runbooks, coordinate incident response, communicate with customers, manage maintenance windows, resolve Tier 1 tickets, and improve monitoring and operational procedures. The role provides 24/7 rotating coverage, including nights, weekends, and holidays, while maintaining shift handoffs and active-incident documentation.