SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 180,000 - 220,000 / annual
Bedrock Ocean builds and operates autonomous underwater vehicles (AUVs) that collect georeferenced ocean-floor data at commercial scale. The company is scaling toward continuous, around-the-clock data collection campaigns and is building AI agents on Amazon Bedrock to support ocean data, internal operations, and customer platforms.
You will lead the design and implementation of the AI infrastructure that powers these agents. This role goes beyond building agents on existing platforms—you will create the infrastructure itself, including the orchestration layer, the data and retrieval pipeline, and the security model required to work with production data. You will also build the tools and abstractions that allow the engineering team to implement AI features independently.
Key responsibilities include:
- Architect the Amazon Bedrock integration, including agent and action group configuration, backend APIs, model access, throughput, and cross-environment deployment.
- Own the end-to-end retrieval pipeline from ingestion and chunking to embedding and storage in Amazon OpenSearch Serverless, optimizing for index design, cost, and capacity.
- Adapt ingestion pipelines for internal knowledge, ocean data, and customer platforms, addressing challenges specific to geospatial and large-binary datasets.
- Implement robust security including Bedrock Guardrails, VPC and PrivateLink network boundaries, least-privilege IAM, and audit trails to ensure data isolation.
- Build mechanisms to enforce approval boundaries for autonomous actions, ensuring agents are safe and monitored.
- Establish comprehensive LLMOps and observability using CloudWatch and LLM-specific tools like Langfuse or Phoenix.
- Create infrastructure to run automated evaluations, track results, and manage release gates for model accuracy.
- Provide the team with abstraction layers, SDKs, and self-service environments that enable independent AI feature development.
- Manage the environment as code across all stages, ensuring deployment safety and participating in incident reviews.
This position combines software engineering, data engineering, and infrastructure operations. Security is core to the role—you will define operational boundaries for agents with tool access, including what they can access, the actions they can perform autonomously, and the monitoring required to detect issues. You will play a key role in defining the rollout sequence, prioritizing internal engineering and operational systems first, followed by ocean and survey data products, with customer-facing retrieval as the final, high-stakes phase.
REQUIREMENTS:
- 8+ years in software and infrastructure engineering, including deep production backend experience (Python or TypeScript preferred, Go acceptable) and staff-level ownership of technical direction.
- Hands-on experience standing up Amazon Bedrock in production: agents, knowledge bases, guardrails, model access, and throughput/quota decisions.
- Containerized service deployment on ECS, EKS, or Lambda, with CI/CD you have owned rather than inherited.
- Practical RAG and vector search experience: embeddings, chunking strategies, semantic search quality, and operating a managed vector database (OpenSearch Serverless, Pinecone, pgvector, or similar) at production scale and cost.
- Real data engineering: built or substantially extended ingestion pipelines over messy, heterogeneous, unstructured sources; think about freshness and correctness as SLAs.
- Strong AWS ecosystem expertise: IAM roles and least privilege for machine identities, VPC networking and PrivateLink, Lambda, S3, KMS, CloudWatch, and infrastructure as code (Terraform, CDK, or CloudFormation).
- Production LLM exposure: moved LLM features or autonomous agents past the prototype stage into environments other people depend on.
- Working point of view on securing agentic systems: scoping tool permissions, prompt injection and exfiltration risk, sensitive data handling in retrieval, and human-in-the-loop placement.
- Experience designing developer-facing APIs, SDKs, or platform services with an API-first mindset.
- Experience building and operating multi-tenant services with isolation guarantees for customer data.
- Platform instinct: build abstractions other engineers stand on; measure success by what they ship.
- Demonstrated technical leadership and system design judgment at staff level: driven architectural direction across pods or teams you do not manage, made it stick through influence.
- Pragmatic builder's bias: reach for boring, fully managed infrastructure before complex self-hosted alternatives.
- Comfort wearing several hats on a small team and discipline to document systems for operation without you.
NICE TO HAVE:
- Experience deploying LLM evaluations to measure accuracy over time and treating eval results as a release gate.
- Involvement in AI red-teaming or the AI security community.
- Experience with GraphRAG or knowledge graphs.
- Experience running retrieval over geospatial, scientific, or large-binary datasets.
- Experience moving data across intermittent or unreliable links: store-and-forward, hub-and-spoke topologies, edge system offload, and reconciliation.
- Compliance experience such as SOC 2, or handling government or defense customer data.
- Background supporting data platforms, autonomous systems, or field operations.
