СШАData Infra - Telemetry and Observability (IC)
1 мес. назад
СШАMiddleOnsite
telemetryalertingmonitoringgpu utilizationcheckpointingdata loadingdistributed training
Individual contributor role to build telemetry and observability systems for datasets, training runs, models, environments, agents, and missions as one AI system.
О компании
- Welcome to Matter, where we are building the future of vision AI: pairing a world-first sensor that sees molecular chemistry, temperature, and 3D shape with a Large World Model that will be the most powerful intelligence engine for the physical world. This system doesn't just see what something looks like; it understands everything from a single pixel. We call this Superintelligent Vision.
- Our team has delivered technologies to Mars for NASA/JPL, designed advanced sensors for U.S. Defense, and built core infrastructure at OpenAI. We are now building the next generation of space- and airborne-based sensing systems.
Обязанности
- Matter is hiring a Telemetry and Observability Engineer to make datasets, training runs, models, environments, agents, and missions observable as one AI system. Reporting to Ignacio Cases Martin, this individual contributor will connect flight and mission reality with infrastructure, model, and product behavior through shared telemetry, traces, alerts, reliability mechanisms, and operational context.
- Build a common telemetry model linking mission state, datasets, code, checkpoints, accelerators, model and prompt versions, environment episodes, agent steps, tools, decisions, cost, latency, and feedback.
- Instrument distributed training, data loading, GPU utilization, checkpointing, experiment health, batch and online inference, and model-serving behavior.
- Instrument agent workflows across prompts, context, retrieval, memory, planning, tool calls, graph state, human intervention, evidence, outcomes, and safety controls.
- Connect planned-versus-actual mission state, hardware context, time, geometry, and geolocation to relevant datasets and model results.
- Establish actionable service objectives, alerts, incident interfaces, rollout and rollback signals, and escalation paths across platform, model, environment, and agent failures.
- Build capacity and cost visibility across storage, processing, accelerators, training, evaluation, inference, retrieval, and agent execution.
- Teams can correlate mission, data, infrastructure, model, environment, and agent behavior during normal operation and incidents.
- Alerts and service objectives identify actionable failures without obscuring scientific or operational context.
- Capacity, cost, version, and reliability signals support safe deployments and faster diagnosis across the AI stack.
Требования
- To comply with U.S. export regulations, applicants must be one of the following:
- A U.S. citizen or national
- A lawful permanent resident (green card holder)
- Eligible to obtain required authorizations from the U.S. Department of State
Будет плюсом
- Experience with MLOps, GPU workloads, model serving, agent observability, reinforcement-learning environments, mission telemetry, or scientific instrumentation.
- Experience operating high-throughput inference, streaming systems, geospatial pipelines, or mixed cloud and edge deployments.
- Experience with reliability platforms or observability products used by multiple engineering teams.
- Familiarity with aerospace, defense, regulated operations, or other settings requiring formal change control and incident evidence.
Условия
- This role is based in San Francisco, CA, and requires onsite work.
- At Matter, we believe in rewarding high performance and providing the support you need to thrive. Our compensation and benefits package includes:
- Competitive compensation based on experience
- Early-stage equity package
- 100% employer-paid health, dental, and vision coverage
- Opportunity to work on novel sensing, data, and AI systems with real-world deployment paths to the largest industries in the world
Другое
- Experience with ML observability, telemetry, time-series systems, distributed training or inference, cloud infrastructure, or production AI platforms.
- Strong hands-on engineering skills in observability, automation, APIs, infrastructure, incident tooling, and systems debugging.
- Understanding of metrics, logs, traces, events, model and data monitoring, service objectives, capacity, release safety, and incident diagnosis.
- Ability to reason about AI failure modes including bad datasets, training instability, stale checkpoints, drift, version mismatch, environment bugs, and agent-tool failure.
- Experience designing telemetry schemas and correlation workflows across multiple systems, time domains, and ownership boundaries.