What you'll be doing:
Architect and build an agentic AIOps system that autonomously monitors GPU fleet health, aggregates and correlates massive telemetry streams, surfaces intelligent alerts, and orchestrates multi-step diagnostic workflows and corrective actions – powering real-time dashboards, automated root-cause analysis, and proactive incident response.
Research, evaluate, and prototype data storage strategies and data representations across diverse database technologies and modalities, ensuring AI models are trained on high-quality, well-structured data that improves predictive accuracy and generalization.
High-Scale Engineering: Design distributed systems to handle the extreme telemetry density of large-scale AI clusters, ensuring efficient data ingestion, processing, and real-time analysis.
Instrument services with deep observability (metrics, logs, traces) to support rapid debugging and continuous performance improvement.
Build and own the model-serving infrastructure that operationalizes predictive algorithms at scale – packaging, versioning, deploying, and monitoring AI models in both SaaS and on-premises environments.
Contribute to the platform's core libraries and abstractions that accelerate development across the broader AIOps engineering team.
What we need to see:
B.Sc./M.Sc. in Computer Science, Computer Engineering, or a related technical field.
8+ years of software engineering experience building production distributed systems.
Core Systems Programming: Expert-level proficiency in languages such as Go, C++, or Rust, with a focus on high-performance, concurrent architectures.
Solid understanding of Kubernetes and container-based deployments for production services.
Experience deploying, monitoring, and maintaining ML models or data-intensive services in a production environment.
Comfort working in ambiguous, fast-moving environments where the product is still being shaped.
Ways to stand out from the crowd:
Experience building ML model-serving platforms or MLOps tooling (model registries, A/B rollout frameworks, feature stores) at scale.
A track record of taking systems from prototype to stable, production-grade platform serving real enterprise customers.
A "Systems" Thinker: You don't just write software; you understand the full stack, from how data moves across the wire to how its processed in a distributed cluster.
Practical Innovation: The ability to simplify complex problems and build internal tools or frameworks that empower other engineering teams to move faster.
