AI Engineer - Agent Infrastructure & AI Workflow
New
T
TribalScaleAI infrastructure
Remote USAFull-TimeMiddle
Salary not disclosed
Apply NowOpens the employer's application page
Job Details
- Required Skills
- AWSKubernetesAWS Lambda
Requirements
- Have hands-on experience with AWS services including Bedrock, Lambda, ECS or EKS, S3, and Step Functions.
- Have experience with GenAI infrastructure, including LLM integration, agent frameworks, MCP, semantic caching, and LLM evaluation.
- Be familiar with Amazon EKS and Kubernetes-based deployments.
- Be familiar with GitHub or GitLab workflows, secrets management, and secure deployment practices.
- Have knowledge of observability tools such as Datadog and log aggregation strategies.
- Be able to write clean, maintainable code and contribute to shared documentation and best practices.
- Have experience collaborating directly with customers, stakeholders, or cross-functional delivery teams.
- Be able to explain AI-system behavior, limitations, risks, and technical trade-offs clearly.
- Be comfortable working in fast-moving environments where requirements and technical constraints evolve.
- Be able to travel occasionally for client workshops or on-site delivery, where required.
Responsibilities
- Design and implement orchestration for long-running agent pipelines on AWS, including job orchestration, artifact and manifest generation, checkpointing, retries, and prompt and model versioning.
- Run multi-model inference across Amazon Bedrock and self-hosted models, with routing, an independent judge model, semantic caching, and token-cost and latency controls.
- Provide shared evaluation and tracing hooks so Model Delivery can measure pipeline stages and compare models, prompts, and pipeline versions.
- Build and operate the internal AI gateway, agent runtime, and tool integrations for AI-enabled software development workflows.
- Develop and maintain CI/CD pipelines and infrastructure as code using Terraform or CDK, with secure secrets management and least-privilege access.
- Instrument agents and pipelines, and maintain dashboards, alerts, and runbooks for success rate, cost, latency, and failure modes.
- Work with product teams that own customer-facing MCP and API surfaces to keep internal agent tools and product interfaces consistent.
View Full Description & ApplyYou'll be redirected to the employer's site