Enterprise AI Systems Engineer
About this role
Juul Labs is building centralized enterprise AI systems and agentic platforms to support company-wide adoption of tools like Claude, Gemini, and Cursor. This role centers on owning the architecture end-to-end, deciding how AI runs internally, and operating the LLM gateways, Model Context Protocol servers, and cloud infrastructure on AWS and GCP. Candidates will build integrations, manage tenant security and provisioning, and partner with cybersecurity to enforce data governance across a growing user base.
The position exists to scale AI enablement, track consumption and spend, and build reliable automation pipelines using Python and TypeScript. Someone who thrives in a hands-on technical environment, takes ownership of complex systems without inheriting pre-existing designs, and communicates clearly with both engineers and finance stakeholders will do well here.
What you will do
- Administer enterprise AI and SaaS tenants including user provisioning, workspace configuration, and tenant security settings.
- Build and maintain platform content such as skills, prompts, projects, connectors, and internal documentation.
- Monitor platform availability, triage user issues, and manage escalations while tracking consumption and spend across services.
- Build and operate Model Context Protocol servers and LLM gateways that handle routing, authentication, rate limiting, and policy enforcement.
- Architect and run AI workloads on AWS and GCP using compute, networking, IAM, secrets management, Bedrock, and Vertex AI.
- Write production code primarily in Python and TypeScript using GitHub for version control, actions pipelines, and branch protection.
- Collaborate with cybersecurity and AI governance teams to enforce data handling, retention, and access policies.
What you need
- Five or more years of experience in software, cloud, or platform engineering with recent hands-on work building LLM-based systems.
- Demonstrated experience administering enterprise SaaS or AI tenants at scale, including access management and cost control.
- Working knowledge of commercial and open-weight LLMs covering prompt engineering, evaluation, retrieval, and agentic patterns.
- Hands-on experience with Model Context Protocol, MCP gateways, or equivalent API integration and middleware.
- Proficiency with AWS and GCP architecture, including IAM, networking, secrets management, and managed AI services.
- Strong software engineering fundamentals using GitHub, version control, code review, testing, CI/CD, and infrastructure as code.
- Strong programming skills in Python and JavaScript or TypeScript for building internal tooling and automation.
- Understanding of security and data governance principles such as least privilege, data classification, and auditability.
Nice to have
- Vertex AI or Amazon Bedrock
- Self-hosted open-weight model deployment
- Vector databases and RAG pipelines
- Splunk or another SIEM
- Cloud cost management
Benefits
- A place to grow your career and set big goals
- Equity and performance bonuses
- Cell phone subsidy, commuter benefits, and discounts on JUUL products
- Medical, dental, vision, disability, and life insurance, plus family support, wellness, legal, and employee assistance program benefits
- 401(k) plan with company matching
- Biannual discretionary performance bonuses
Skills and tech stack
Questions candidates ask
What locations are eligible for this role?
The position supports multiple tiers of locations across the United States, including major metropolitan areas and regional hubs, with salary ranges adjusted based on geographic labor costs.
What compensation and bonus structure is included?
The role offers a salary range depending on location and level, alongside equity, biannual discretionary performance bonuses, and a 401(k) plan with company matching.
What are the primary technical tools and platforms used?
Engineers work with enterprise tenants like Claude, Gemini, and Cursor, build on AWS and GCP using Bedrock and Vertex AI, and write code primarily in Python and TypeScript.