SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Beacon AI is building an AI platform to make flying safer, more efficient, and more capable, backed by top investors with a dozen Department of Defense contracts and partnerships with major airlines. The company operates as small, focused teams that own what they build, ship quickly, and learn fast.
You will design, implement, and operate cloud infrastructure and LLM platform services that are scalable, reliable, and secure. Key responsibilities include designing and maintaining AWS infrastructure using Infrastructure as Code (CDK or Terraform), building CI/CD pipelines with GitHub Actions and CodePipeline, and operating secure networking with VPCs and IAM controls.
On the LLM platform side, you'll stand up and operate model endpoints using AWS Bedrock and SageMaker, build application services that call LLMs through clean APIs with streaming and batching strategies, and implement prompt and tool execution flows with LangChain. You'll design chunking and embedding pipelines for documents and multimedia, operate vector search systems using OpenSearch Serverless or pgvector, and build knowledge bases with data syncs from multiple sources.
You'll create evaluation harnesses for prompts and retrievers, instrument telemetry with CloudWatch and OpenTelemetry, build cost dashboards with budgets and alerts, and implement safety guardrails including PII detection and access controls. Additional responsibilities include building data ingestion and processing pipelines, managing IoT deployment infrastructure for edge devices, and optimizing inference performance through model selection and autoscaling.
Success requires end-to-end ownership from design through production, hands-on experience shipping LLM-powered applications, deep AWS knowledge across core services, comfort building data pipelines in Python, and a security-first mindset. You'll communicate tradeoffs clearly and align partners across product, security, and engineering teams. Bonus experience includes serverless/container platforms, vector databases at scale, Bedrock Guardrails, GPU workloads, and aviation domain knowledge.
