Data & AI Engineer - (Hybrid)
Indexed description
Job Summary
The Data & AI Engineer designs, builds, and operates the data and AI systems behind Cruise Planners' analytics, advisor-facing products, and internal automation. The role combines production data engineering on AWS (Glue, Redshift, Step Functions, Lambda, S3) with applied generative-AI work on Amazon Bedrock (Data Automation, Knowledge Bases, AgentCore) and the Strands Agents SDK.
Day-to-day development is conducted through agentic coding tooling — Claude Code with MCP — used as a primary engineering surface, not as autocomplete, with the pace and scope of delivery expected to exceed what traditional tooling supports. The team is small: engineers scope, build, deploy, document, and operate their own work end to end.
The items listed below are intended to provide an overview of the essential functions of the job. This is not an exhaustive list of all functions and responsibilities that the position may be required to perform.
Responsibilities
- Design and operate data pipelines on AWS — Glue, Step Functions, Lambda, EventBridge, S3 — feeding the Redshift data warehouse.
- Stand up and operate containerized production pipelines on AWS Fargate / ECS (scheduled ingest or scrape, enrich, deliver) with CloudWatch alarms and health monitoring.
- Design dimensional data models — facts, conformed dimensions, and analytics marts — on Redshift over a zero-ETL replica of the operational source.
- Implement document-extraction pipelines using Amazon Bedrock Data Automation blueprints and deliver structured output to downstream consumers via webhook or queue.
- Build and operate Bedrock Knowledge Bases for retrieval-augmented generation: source preparation, indexing, retrieval evaluation, and cost / latency tuning.
- Build agents using the Strands Agents SDK on Bedrock AgentCore — supervisor / sub-agent topologies, tool definitions, and deployment hardening.
- Build applied-LLM data products for reporting, search, and agentic operations.
- Deploy and operate services on AWS end to end, including internal web apps and demos.
- Build internal tooling and integrations, including MCP servers.
- Prototype and ship new AI-powered applications for advisors and internal teams.
- Maintain operational quality — logs, alarms, runbooks, on-call response, post-incident notes — and continuously monitor quality and compliance with data-privacy regulations.
- Conduct day-to-day development through Claude Code with project-scoped MCP servers, document decisions and architectures in the team wiki.
- AWS data and serverless services.
- Amazon Bedrock and AgentCore Runtime.
- Python (primary); SQL (baseline).
- Data warehousing and dimensional modeling.
- Agent and LLM frameworks.
- Infrastructure-as-code.
- Claude Code as the primary development surface, including parallel sessions.
- MCP servers (AWS, Excolo, Atlassian) as deterministic tool routing for work that previously spanned terminal, IDE, and browser.
- Comfort treating the model as a collaborator — generating alternatives, having design conversations, validating choices before committing — rather than as autocomplete.
- Writes documentation a teammate can pick up and act on without a meeting.
- Explains trade-offs in plain language to non-engineers (Marketing, Operations, Finance).
- Strong problem-solving, communication, and collaboration skills.
- Bachelor's degree in Computer Science, Data Science, Statistics, Engineering, or a related field. Master's preferred at the Senior level.
- Relevant production experience scaled to level (Data & AI Engineer I / II / Senior).
- Authorization to work in the United States without sponsorship.
- Data & AI Engineer I — Executes scoped work end to end under guidance; ships well-defined pipelines, blueprints, or Knowledge Base load jobs.
- Data & AI Engineer II — Owns workstreams end to end (design, build, deploy, operate) and makes architecture choices within an established pattern; writes the playbook a peer can ship from.
- Senior Data & AI Engineer — Owns multi-system architecture; stands up a new pattern (e.g., AgentCore deployment, multi-agent topology) without precedent; coaches peers; represents the team in cross-functional design. Operates across the full stack — Fargate production pipelines, inbound AWS deployments, applied-LLM data products, and end-to-end agent architecture.
Create a free Caio profile to unlock more results and save your role and location preferences.
Unlock free search