STAFF, SOFTWARE ENGINEER

What You'll Do

Design and own core platform capabilities — plugin and extension systems, agent orchestration engines, semantic and data abstraction layers — enabling product teams to onboard new AI capabilities with minimal platform involvement

Architect and implement multi-step, parallel, and cross-domain agentic workflows powered by large language models, ensuring correctness, reliability, and safe execution at every step

Build and enforce governed tool execution and security boundaries — access controls, tenant isolation, scope enforcement — so that AI agents operate within declared, auditable constraints

Define and evolve the platform's data and semantic layer, bridging heterogeneous sources (relational, vector, semantic-modelled, API) into a unified, agent-understandable representation shared across all products

Establish and uphold non-functional requirements — latency SLAs, throughput targets, data isolation guarantees, and availability contracts — as first-class platform standards, not afterthoughts

• Own observability and explainability as platform primitives — audit pipelines, structured logging, response attribution, and governance metadata that make AI outputs trustworthy and debugable in production

Own observability and explainability as platform primitives — audit pipelines, structured logging, response attribution, and governance metadata that make AI outputs trustworthy and debugable in production

Define architectural dependency rules and enforce them through code, tests, and design reviews — preventing structural violations before they compound

Partner across engineering, product, and data science teams to translate platform capabilities into reliable, well-documented contracts that internal teams can build against confidently

What You'll Bring

10+ years of software engineering experience, with meaningful depth in platform, infrastructure, or foundational systems — not just application development

• Hands-on experience designing and shipping production LLM-based systems — agent orchestration, tool use, RAG pipelines, multi-step reasoning, or structured output generation

Hands-on experience designing and shipping production LLM-based systems — agent orchestration, tool use, RAG pipelines, multi-step reasoning, or structured output generation

Strong software engineering fundamentals — clean API design, distributed systems patterns, async execution, fault tolerance, and observability

Experience designing extensible plugin or extension systems with stable, versioned contracts and well-defined conformance expectations

Proficiency in Python and familiarity with the AI/ML ecosystem; comfort reasoning about latency, throughput, and cost tradeoffs in inference-heavy workloads

Deep appreciation for architectural principles — separation of concerns, dependency inversion, layered design, and enforcement of contracts through code rather than convention

Preferred

Experience with LLM governance, AI safety frameworks, guardrails, or responsible AI patterns in production

Background in semantic modelling, data mesh, or building canonical data layers consumed by downstream AI or analytics systems

Prior work in a platform engineering role serving multiple product teams as internal customers

Familiarity with agentic frameworks (LangGraph, LangFuse, GoogleADK, CrewAI, or equivalent) and the tradeoffs of building on top of vs. owning the orchestration layer

Minimum Qualifications...

• Outlined below are the required minimum qualifications for this position. If none are listed, there are no minimum qualifications.

Preferred Qualifications...

• Outlined below are the optional preferred qualifications for this position. If none are listed, there are no preferred qualifications.