3 months ago
Remote, United StatesSenior
Responsibilities
- Build, deploy, extend, and operate internal AI chat, agent, retrieval, and developer-workflow systems.
- Develop GitLab-integrated agents that create merge requests, respond to reviews, and iterate through existing engineering workflows.
- Design secure sandboxes and isolated environments for agent code execution, repository access, tools, and hardware-in-the-loop use cases.
- Own the AI gateway layer, including model routing, credential management, access policy, telemetry, and cost attribution.
- Operate the AI stack through monitoring, upgrades, incident response, and continuous improvement.
- Partner with security and compliance stakeholders to implement CUI handling, data residency, provider selection, audit logging, and other boundary requirements.
- Produce architecture, data-flow diagrams, and logging evidence supporting compliance documentation.
- Evaluate applied-AI tooling, define the roadmap, and communicate technical opportunities and limitations to leadership.
Requirements
- Several years of experience building, shipping, and operating production software.
- Fluency in at least one of Python, Go, or TypeScript, plus Terraform.
- Experience with API design, testing, code review, and production service ownership.
- Hands-on experience building LLM-powered systems involving retrieval pipelines, tool use or MCP, agent orchestration, prompt and context management, and evaluation.
- Experience deploying, extending, and integrating open-source applications such as chat frontends, gateways, or developer tools.
- Experience with Git-hosting APIs, webhooks, and CI/CD integrations for developer workflows.
- Cloud and Kubernetes engineering experience with technologies such as containers, Helm, Terraform, GKE, and IAM.
- Understanding of sandboxing, isolation, network policy, and least-privilege credentials for safely running model-generated code.
- Ability to elicit requirements from security stakeholders and design systems that are observable and auditable in regulated environments.
- Clear technical and executive written communication and the autonomy to define and validate a roadmap.
- Preferred: experience in CMMC, DoD Impact Level, FedRAMP, or NIST 800-171 environments.
- Preferred: self-hosted inference with vLLM, TGI, or similar systems, including GPU provisioning and open-weight model deployment.
- Preferred: experience with LLM gateway or proxy layers such as LiteLLM and per-team cost attribution.
- Preferred: hardware-in-the-loop, lab automation, device access control, DLP, GCP services, Bazel or other hermetic build systems, and observability stacks.
Benefits
- Flexible working arrangements, including hybrid remote/in-office schedules.
- Professional development and career advancement opportunities.
- 401(k), dental, vision, health, and life insurance benefits.
- Paid time off and equity options.
- Opportunity to work on aerospace communications and national-security programs.
Tech Stack
Categories
About Aalyria
Aalyria builds networking technologies for operating high-throughput, moving networks across land, sea, air, and space. Its products include Tightbeam, an ultra-high-speed free-space optical communications system, and Spacetime, a platform-as-a-service for modeling, planning, and orchestrating dynamic networks for commercial and government operators. Privately held and headquartered in Livermore, California, Aalyria originated from technologies developed at Alphabet and serves aerospace, telecommunications, and national security markets.