Senior AI Agent Engineer — Production LLM Systems

Jobot; New York (NY); On-site; USD 185,000 - 235,000; Full time

Benefits offered by this job

Equity / ownership

Direct access to founders and CTO

Very high technical ownership

Frontier agentic-AI work

Early-stage impact

Strong founding team

Funded startup

Interesting customer/mission exposure

SoHo / NYC office

Job summary

Jobot in SoHo, New York City, seeks a Staff/Senior Software Engineer specializing in Agent Engineering to advance the agent runtime powering the platform's research, analysis, and action capabilities.

You will own and improve the systems that make agent behavior reliable in production, including execution state, tool use, memory, model routing, failure recovery, evaluation, observability, and security.

Qualifications

• Production orchestration of stateful, long-running, multi-agent workflows.
• Tool and action framework with contracts, permissions, and approval boundaries.
• Retrieval, context engineering, evidence management, and persistent agent memory.

Responsibilities

• Own and advance the platform's agent runtime and foundational systems.
• Build infra to manage long-running work with visibility, steerability, and project linkage.
• Design and build tool/action framework for safe interaction with internal/external systems.
• Connect agents to retrieval/knowledge systems for grounded reasoning and citations.
• Develop persistent memory and context-management for complex workflows.
• Improve evaluation, tracing, and release processes for agent behavior.
• Enhance reliability while managing latency, cost, security, and risk.
• Diagnose and resolve production failures involving nondeterministic AI.

Skills

Production orchestration

Tool infrastructure

Retrieval and memory

Model-runtime engineering

Evaluation and observability

LLM behavior reliability

Long-running multi-agent workflows

Job description

Jobot in SoHo, New York City, seeks a Staff/Senior Software Engineer specializing in Agent Engineering to advance the agent runtime powering the platform's research, analysis, and action capabilities.

You will own and improve the systems that make agent behavior reliable in production, including execution state, tool use, memory, model routing, failure recovery, evaluation, observability, and security.