Senior AI / LLM Engineer (Agentic RAG & LLMOps)

Our Client - We're pioneering a fundamental shift in cybersecurity—moving organizations from fragmented, reactive defense to unified, proactive protection. Our AI-powered platform synthesizes intelligence from 150+ disparate security tools, transforming overwhelming noise into crystal-clear risk prioritization through breakthrough predictive technology built on 25+ patents in breach path prediction and threat analysis. Founded by proven security innovators with track records of building, patenting, and successfully exiting industry-leading companies, we're solving the problem that keeps CISOs awake: understanding what truly threatens your organization amid endless alerts. We're building an exposure-centric security mesh that continuously optimizes defenses and shrinks attack surfaces using patented intelligence that predicts and prevents breaches before attackers strike. Join us to architect the future of enterprise security—where deep technical innovation meets battle-tested expertise, turning complexity into clarity and reaction into foresight.

We’re hiring a Senior AI / LLM Engineer to own our agentic RAG, text-to-SQL copilots,

and LLMOps systems end-to-end—from architecture and orchestration to evaluation,

guardrails, and high-scale production operations.

You will design reliable, high-accuracy AI systems that power mission-critical workflows,

while driving the engineering standards, tooling, and infrastructure that make them

scalable.

What You’ll Do

● Design and scale agentic RAG and text-to-SQL copilots capable of handling

50K+ daily queries with 99.9%+ reliability and high semantic accuracy.

● Build, maintain, and optimize our LLMOps stack using tools like LangGraph,

LangSmith, MLflow, Kubernetes, async inference, and cloud LLM providers such

as AWS Bedrock, Google Vertex, Azure OpenAI, Anthropic, etc.

● Develop and maintain MCP server integrations, ensuring robust and efficient

runtime execution across agents and tools.

● Implement evaluation frameworks and guardrails (including AI-as-a-Judge,

safety filters, grounding checks) to minimize hallucinations, reduce drift, and cut

token/cost overhead by ~30%.

● Own system observability & performance, including latency, throughput, cost

optimization, caching, and retrieval quality.● Optimize inference, retrieval, and orchestration pipelines for scale and reliability.

● Work with product, infra, and leadership teams to define SLAs, unblock customer

needs, and deliver enterprise-grade features.

● Use AI-assisted development tooling (GitHub Copilot, MCP-enabled IDEs, Claude,

GPT, etc.) to accelerate development velocity and quality.

What We’re Looking For

● 5+ years in software or ML engineering, including production-grade LLM or RAG

systems.

● Strong Python engineering skills and hands-on experience with RAG, agent

architectures, tool-calling, and text-to-SQL copilots.

● Proven experience with MCP servers, vector databases, and

retrieval-augmented architectures.

● Expertise in agent development, LLM integration workflows, prompt engineering,

and runtime orchestration systems.

● Hands-on with container orchestration and infra: Kubernetes, async workers,

queueing systems, observability stacks, etc.

● Experience setting up LLM evaluation pipelines, guardrails, monitoring,

experiment tracking, and regression testing.

● Experience with multiple Agent SDKs including:

○ Anthropic SDK

○ ClaudeAgent SDK

○ Google ADK (Agent Developer Kit)

○ (Plus bonus: LangChain, LlamaIndex, AutoGen, or custom agent runtimes)● Ownership mindset with the ability to convert prototypes into robust, high-traffic

production systems.

Write to sanish@careerxperts.com to get connected!