Member of Technical Staff, Agentic Systems
San Francisco • FullTime
Posted 13d ago
Remote Work Policy
On-site
Employment Type
FullTime
Categories
AI Agent Engineer
About the job
Mercor is seeking a Member of Technical Staff to build the backend systems and agent infrastructure for their talent network. A significant portion of Mercor's operations rely on agents for sourcing, screening, matching, and supporting experts with minimal human intervention. This role involves designing and implementing the services these agents run on, including orchestration, tool interfaces, evaluation, and auditable data models. It's a hands-on engineering position where you will write code, own systems end-to-end, and collaborate with product, operations, and AI researchers to define agent capabilities and human involvement.
Responsibilities
- Design, build, and own backend services for agentic workflows, including orchestration, tool calling, retries, fallbacks, and human-in-the-loop review.
- Deploy agents to production and iterate based on real usage, optimizing prompts, tools, guardrails, cost, and latency.
- Develop evaluation infrastructure for assessing agent performance, including offline evaluations, online metrics, and regression detection.
- Design scalable core services, data models, and APIs that accommodate network growth and evolving agent behavior.
- Ensure agent decisions are observable, auditable, and reversible for quick failure detection and understanding.
- Collaborate with product, operations, and AI researchers to determine agent responsibilities, deterministic processes, and necessary human oversight.
- Establish backend engineering standards and patterns, and mentor other engineers.
Requirements
- 5+ years of professional backend engineering experience.
- Strong fundamentals in distributed systems, service architecture, data modeling, and scalable API design.
- Hands-on experience building production systems with LLMs, such as agents, tool use, retrieval, or evaluation systems, or a demonstrated ability to quickly learn and ship new systems.
- Comfort with ambiguity and the unique failure modes of agent systems.
- Proven track record of owning systems end-to-end, from design to production operation.
- Proficiency with modern AI development tools like Cursor, Claude Code, or GitHub Copilot.
- Excellent communication skills, with the ability to articulate complex tradeoffs clearly.
- High ownership, pragmatism, and a bias towards shipping.
Benefits
- Generous equity grant vested over 4 years
- $20K relocation bonus (for Bay Area moves)
- $10K housing bonus (for living within 0.5 miles of the office)
- $1.5K monthly stipend for meals
- Free Equinox membership
- Health insurance