Drimva

Artisan AI

Staff AI Engineer - Agent Architecture & Behavior

$250K–$325K

San Francisco · On-site · Staff · Full-time · Posted Sep 13, 2026

Education
No required degree; no required publication record, previous employer, or agent framework.
Visa sponsorship
US visa sponsorship available.

About the job

Designs and builds the core AI architecture for an agentic product, from agent behavior and multi-agent coordination to evaluation and reliable production execution.

What you'll do

  • Design agent execution, planning, tool use, verification, context, and multi-agent coordination.
  • Build resilient stateful workflows, evaluations, and experiments; use results to improve quality, reliability, latency, and cost.
  • Own AI architecture and technical standards, make model and tooling decisions, and work with product and infrastructure engineers to ship production services.

What they ask for

  • Personally built and shipped a substantial agentic system, through production use or rigorous, reproducible open-source work.
  • Deep practical experience with LLM tool use, planning, context engineering, evaluations, and multi-agent coordination or substantial parallel agent/tool execution.
  • Hands-on experience with browser or computer automation in an agentic system, including state observation, effect verification, and failure recovery.
  • Strong software engineering in Python, TypeScript, or a comparable language; comfortable with asynchronous services, state machines, persistence, concurrency, retries, and cancellation.

Nice to have

  • Experience in agent memory and retrieval, skill acquisition, reinforcement learning or post-training, trajectory datasets, sandboxed execution, distributed systems, inference optimization, or multimodal and voice models.

Pay

$250,000–$325,000 USD annually (base salary)

Tools

Python, TypeScript

AI focus

Builds agentic AI architecture, including agent behavior, multi-agent systems, tool use, context and memory, evaluation, and dependable execution.

Model-written summary · the original posting is authoritative

First seen Oct 29, 2026 · Last verified Nov 5, 2026 (UTC)