Staff Engineer, AI R&D

Act as the technical lead of AI initiatives, from the first definition of the problem through delivery in production, working alongside the engineers and team leads who execute them.

Work directly with peers and business stakeholders to understand needs, agree on scope and priorities, and keep expectations aligned as the work evolves.

Translate business goals into technical plans, milestones and trade-offs, and raise risks early.

Make sure initiatives reach production with outcomes that can be measured and explained to the business.

Lead the design and architecture of AI systems, including agentic workflows, retrieval, tool use, orchestration and state management, built by one or several teams.

Define reference architectures, patterns and engineering standards that other teams adopt and reuse.

Make and document trade-off decisions across quality, latency, cost, security and vendor dependency.

Solve the most complex technical problems in your area and unblock the teams around you.

Establish how AI behavior is measured: evaluation datasets, regression tracking, quality metrics and acceptance criteria for agents and LLM-based features.

Ensure systems are tested against realistic cases, not only demos, before and after they reach production.

Define the telemetry needed to understand cost, latency, quality and outcomes end to end.

Lead operational practices: reliability, scalability, resilience, observability and incident learning for AI services.

Examine production behavior, identify structural improvements and drive them to completion with the owning teams.

Keep cost under control through routing, caching, context management and capacity decisions.

Design controlled access for agents that act on internal systems: permissions, guardrails, audit trails and human-in-the-loop checkpoints.

Work with Security, Legal, Compliance and IT so that governance requirements are built into reusable components and not rebuilt for every use case.

Mentor engineers and tech leads, and raise the technical level of the teams you work with.

Turn what you learn into playbooks, templates, shared libraries and documentation that others can use without you.

Communicate clearly with technical and non-technical stakeholders, and align teams around shared standards.

Share knowledge through internal write-ups and tech talks, and occasionally through external meetups and conferences.

8+ years of software engineering experience, including significant experience operating at senior or Staff-level scope.

2+ years of hands-on experience building and operating systems based on LLMs, ideally in production.

Experience designing agentic or multi-step AI systems involving tool use, orchestration, state, retrieval or external integrations.

Strong foundations in distributed systems, including synchronous and asynchronous communication, and in software architecture principles and practices.

Solid knowledge of cloud infrastructure, preferably AWS, and of running services securely and cost-consciously.

Experience with observability, testing and evaluation of complex systems.

Track record of technically leading initiatives that involve multiple teams, from design to production.

Comfortable working directly with business stakeholders and engineering peers to agree on scope, priorities and outcomes.

Ability to understand trade-offs, make decisions with incomplete information and stay effective in crisis situations.

Experience defining engineering standards and best practices that teams actually adopt.

Able to influence without authority, and to explain technical decisions to non-specialists in concrete, concise terms.

Experience with agent frameworks and protocols such as LangChain, LangGraph, the Claude Agent SDK or MCP.

Experience with Kubernetes and containerized execution environments.

Experience with LLM gateways, model routing and cost management.

Background in machine learning or applied research.

Experience in regulated environments such as payments or financial services.

Builder attitude: you prefer reusable platforms and tools over one-off solutions.

Curious and biased toward experimentation, combined with disciplined measurement and risk awareness.

Comfortable with ambiguity in a fast-moving field, and able to structure your own work and keep stakeholders informed.