Quick Summary
Lead the technical strategy and hands-on execution for building production-grade multi-agent systems and security chatbots. Scale Our Capabilities: Act as an accelerator for the team,
What We Offer
~2 min readSwimlane is redefining security operations with Agentic AI automation that empowers organizations to work smarter, respond faster, and stay ahead of threats. Our low-code platform combines automation, orchestration, and intelligent reasoning to unlock true operational autonomy across the modern SOC.
At Swimlane, we put people first. We foster a culture of innovation, trust, and continuous improvement—where your ideas matter and your work drives meaningful change. Join us and help build the next era of Agentic AI-powered security operations.
Senior AI Systems Engineer
Agentic AI & AI SOC
About the Role
~1 min readOur current team is firing on all cylinders to deliver core features, but as our product scope rapidly expands, we need a technical driver to help us scale. We are looking for a Principal AI Systems. Engineer to act as a pathfinder and shape the next generation of AI SOC capabilities in the Swimlane Turbine platform.
This position sits within Data Science but is fundamentally a software engineering and integration role—not a model-training role. You will take the wheel on applied LLM projects, driving the transition from single-prompt features to robust multi-agent systems and intelligent customer-facing security chatbots.
Responsibilities
~1 min read- →Drive Architectural Change: Lead the technical strategy and hands-on execution for building production-grade multi-agent systems and security chatbots.
- →Scale Our Capabilities: Act as an accelerator for the team, taking ownership of new AI feature development so we can keep pace with rapidly growing product demands.
- →Build Applied LLM Workflows: Design secure, maintainable workflows using for AI SOC use cases like alert triage, summarization, and workflow automation.
- →Own Evaluation (Evals): Create test sets, define success metrics (accuracy, faithfulness, latency), and run regression tests before and after changes for your features.
- →Monitor Production Behavior: Debug hallucinations and systematically reduce failure modes through better retrieval, prompting, and guardrails (not just "tweak the temperature").
- →Establish Engineering Standards: Define robust patterns for context engineering, tool design, retrieval, and harness engineering to ensure our agents are rigorously evaluated for quality and safety.●
- →Champion Security-First AI: Ensure data privacy, access controls, and prompt injection defenses are embedded deeply into our multi-agent architectures.
- →Collaborate Cross-Functionally: Work seamlessly across Data Science, Engineering, Product, and Platform teams to guide emerging AI capabilities from prototype to production.
Location & Eligibility
About this job
- Who can apply
- India
- Posted
- 24 days ago
Listing Details
- First seen
- September 26, 2026
- Last seen
- October 20, 2026
Posting Health
- Days active
- 23
- Repost count
- 0
- Trust Level
- 20%
- Scored at
- October 20, 2026
Signal breakdown
Stay ahead of the market
Get the latest job openings, salary trends, and hiring insights delivered to your inbox every week.
No spam. Unsubscribe at any time.