Matter is hiring a Product Intelligence Engineer to turn physics-informed models, general-purpose AI, and ultraspectral data into secure, evidence-backed intelligence products. Reporting to Ignacio Cases Martin, this hands-on individual contributor will build agentic workflows, retrieval and tool integrations, reports, and human-in-the-loop experiences that help customers make operational decisions.
Build production workflows that combine physics-informed models, LLMs, VLMs, world models, scientific code, data services, and deterministic rules through stable interfaces.
Design durable execution across state, planning, model routing, retrieval, memory, tool use, retries, timeouts, checkpoints, human approval, and recovery.
Create evidence-backed outputs that distinguish source data, model inference, system-generated synthesis, user input, and unresolved uncertainty.
Partner with product and design to build workflows for review, correction, comparison, approval, reporting, and recovery.
Build context, retrieval, memory, and indexing patterns that respect permissions, freshness, versioning, retention, and evidence lineage.
Evaluate real system behavior across task success, groundedness, scientific validity, safety, latency, cost, and human intervention.
Qualification
Experience with scientificExperience with agent evaluationWhat Success Looks LikeCustomers receive timely
Required
Experience building production AI applications, agent systems, workflow engines, backend services, or complex software products.
Strong hands-on engineering skills across APIs, stateful services, databases, asynchronous execution, testing, debugging, and production operation.
Practical experience with LLMs or multimodal systems, retrieval, memory, tool use, model routing, structured generation, and common failure modes.
Product judgment and the ability to connect technical design choices to the user decision or workflow being improved.
Experience building systems with security, permissions, data isolation, auditability, or human-approval requirements.
Preferred
Experience with scientific, geospatial, industrial, defense, autonomy, or other high-consequence intelligence systems.
Experience building multimodal applications that combine imagery, maps, time series, documents, structured data, or sensor streams.
Experience with agent evaluation, constrained generation, program synthesis, durable workflow systems, or mixed cloud and edge architectures.
Experience working closely with product and design teams to ship customer-facing capabilities.
What Success Looks Like
Customers receive timely, evidence-backed answers and workflows that support real operational decisions.
Agentic capabilities behave predictably across failures, version changes, unavailable dependencies, and human-review boundaries.
Reusable product-intelligence components reduce one-off integration work without obscuring user control or scientific evidence.
