Software Engineer, AI Platform
Aalyria
- San Francisco, California, United States
- Remote, Onsite, Hybrid
- $185,000 - $215,000 a year
- Posted Aug 1, 2026
Job description
About the role
The software engineer will build the AI layer of Aalyria’s engineering organization, designing, shipping and operating internal AI products such as chat and agent platforms, retrieval‑augmented tools, AI‑integrated developer workflows and sandboxed execution environments on Google Cloud (GKE, Vertex AI, GitLab), while navigating a regulated (CMMC, DoD, FedRAMP) environment and owning the end‑to‑end lifecycle.
About the company
Aalyria is a leading technology company that provides laser communications technology and temporospatial software‑defined networking platforms to the aerospace industry, leveraging Google‑acquired tech to pioneer satellite, airborne mesh, cislunar and deep‑space communications and to orchestrate planetary mesh networks across any spectrum, orbit and hardware.
Requirements
- Several years of production software engineering experience with strong fundamentals.
- Fluency in Python, Go or TypeScript and Terraform; experience with API design, testing, code review and owning services.
- Hands‑on experience building LLM‑powered systems including retrieval pipelines, tool use/MCP, agent orchestration and prompt/context management.
- Product sense for internal tooling; experience deploying/extending open‑source chat frontends, gateways or dev tools.
- Developer‑platform integration experience with Git‑hosting APIs, webhooks and CI/CD to embed AI into engineer workflows.
- Cloud and Kubernetes engineering skills: containers, Helm, Terraform, GKE and IAM.
- Sandboxing and isolation literacy for safely running model‑generated code using containers, network policies and least‑privilege credentials.
- Ability to work effectively in a regulated environment, elicit constraints, design observable and auditable systems.
- Clear written communication for technical and executive audiences.
- High autonomy to define and validate own roadmap.
- Preferred: experience in CMMC, DoD Impact Levels, FedRAMP or NIST 800‑171 environments, especially with AI/LLM workloads.
- Preferred: self‑hosted inference experience (vLLM/TGI), GPU provisioning and open‑weight model deployment.
- Preferred: experience with LLM gateway/proxy layers (e.g., LiteLLM) and per‑team cost attribution.
- Preferred: hardware‑in‑the‑loop or lab‑automation experience and device access control.
- Preferred: familiarity with DLP concepts at the model/API layer.
- Preferred: deep Google Cloud Platform expertise (Vertex AI, GKE, IAP, Artifact Registry) and Bazel or similar build systems.
- Preferred: observability stack experience (OpenTelemetry, Grafana/Loki/Mimir/Tempo).