Cybersecurity: Cyber Security Engineer – AI & Agentic Platforms

Make your mark at Comcast -- a Fortune 30 global media and technology company. From the connectivity and platforms we provide, to the content and experiences we create, we reach hundreds of millions of customers, viewers, and guests worldwide. Become part of our award-winning technology team that turns big ideas into cutting‑edge products, platforms, and solutions that our customers love. We create space to innovate, and we recognize, reward, and invest in your ideas, while ensuring you can proudly bring your authentic self to the workplace.

We build and design the tools of tomorrow to stop bad actors from harming our customers and our company. Our mission is to build, maintain, and advance the next generation capabilities that power Comcast’s global cybersecurity operations. This team owns the Premonition platform, which manages, deploys, and composes agentic pipelines at scale—combining large language models, high‑performance data systems, and modern cloud infrastructure to give our security teams a decisive edge.

We work with technologies like Python, SQL, Rust, JavaScript, LLMs (via Ollama, Bedrock, Sage Maker), AWS, EKS, open table formats (Iceberg, Delta, Hudi), and embedded engines such as DuckDB and LanceDB.

As Cyber Security Engineer on the Cybersecurity AI & Data Platform team, you will develop and maintain Premonition, Comcast’s internal platform for managing, deploying, and composing agentic pipelines that power global cybersecurity operations.

• building a robust, scalable platform for LLM and agent‑driven workflows, and
• optimizing the high‑performance data layer supporting those workflows, including storage formats, compression, latency, and query performance across large datasets.

• Design, build, and maintain the Premonition platform for managing, deploying, and monitoring agentic pipelines.
• Develop composable building blocks—agents, tools, pipelines, evaluators, and configuration—to enable rapid workflow assembly and iteration.
• Implement robust APIs, orchestration, and infrastructure integrations on AWS and EKS for reliable LLM and agent‑driven workloads.
• Engineer and optimize the high‑performance data layer, including Iceberg/Delta/Hudi, compression, indexing, latency, and large‑scale query performance.
• Work with embedded engines such as DuckDB and LanceDB to enable interactive, low‑latency analytics.
• Collaborate with cybersecurity stakeholders (SOC, IR, threat hunters, engineers) to translate workflows into secure, automated, observable pipelines.
• Experiment with and product ionize LLM and agent patterns (RAG, tool use, multistep agentic workflows), including evaluation, safety, and guardrails.
• Own the full lifecycle of Premonition services: design, implementation, testing, deployment, observability, performance tuning, and continuous improvement.
• Contribute to and enforce standards and best practices for LLM/agent usage, data management, security, and governance.
• Participate in design/code reviews and foster a supportive, caring, competitive, high‑ownership, deeply technical culture.

• Strong production experience with Python.
• Solid SQL skills and experience working with large datasets.
• Experience designing, building, and operating production APIs or microservices, including testing, observability, CI/CD.
• Experience running workloads on AWS and Kubernetes/EKS.
• Experience with high‑performance data engineering, including:
  • Iceberg, Delta, or Hudi; or
  • optimizing storage, compression, latency, and analytical query performance.
• Ability to work in a deeply technical, high ownership environment and collaborate effectively across functions.
• Strong communication skills with the ability to work closely with security stakeholders and translate workflows into technical designs.

• Demonstrated experience building LLM based applications or agentic workflows (not just prompt tinkering).
• Hands‑on experience with agentic workflows: agents, tools, orchestration, multistep pipelines.
• Experience with LLM platforms:
  Ollama, Amazon Bedrock, Sage Maker.
• Familiarity with orchestration frameworks:
  Lang…