Drimva

Fortinet

Agentic AI Development Engineer

$180K–$260K

Sunnyvale, CA · On-site · Full-time · Posted May 25, 2026

Experience
BS: 5+ years; MS: 3+ years; PhD: 0+ years in the stated field.
Education
BS, MS, or PhD; no equivalent experience alternative stated.
Visa sponsorship
Must be authorized to work in the U.S. without sponsorship.

About the job

Build and scale secure AI-driven operations systems for cybersecurity workflows, spanning backend services, AI agents, real-time data flows, and user-facing experiences.

What you'll do

  • Own complex features from architecture and API design through production delivery and user interactions.
  • Integrate LLMs and generative AI into production workflows, including retrieval, prompt orchestration, fine-tuning, and evaluation.
  • Build scalable, real-time backend and data systems, and deliver trustworthy AI features such as guardrails and human review.

What they ask for

  • Proven experience shipping data-intensive, AI-enhanced applications at scale and owning features through production delivery.
  • With a BS, 5+ years; with an MS, 3+ years; or with a PhD, 0+ years in backend or full-stack systems for high-availability, security-sensitive environments.
  • Proficiency in at least one modern backend language/runtime and framework; strong API, database, authentication, distributed-systems, scalability, and observability skills.
  • Strong problem-solving, communication, and attention to detail; comfortable in ambiguous contexts and with distributed teams.

Nice to have

  • Modern JavaScript/TypeScript and a component-based framework such as React; experience with real-time dashboards and streaming UX.
  • Event-driven architectures, message queues, graph/time-series data, real-time collaboration, or SOC/NOC workflows.

Pay

$179,500–$260,000 USD base salary

Tools

Python, Go, REST, WebSocket, GraphQL, Postgres, JavaScript/TypeScript, React

AI focus

Develop and productionize AI agents and generative AI workflows, including RAG, prompt orchestration, fine-tuning, evaluation, inference optimization, guardrails, and human-in-the-loop review.

Model-written summary · the original posting is authoritative

First seen Oct 28, 2026 · Last verified Nov 4, 2026 (UTC)