Confirmed on the employer's own hiring board on Nov 20, 2026. First seen by Alion on Nov 20, 2026. Mercedes-Benz scores A on the Alion truth index.
Aufgaben
Job Description: AI/ML Engineer (AWS)
About the role
We’re looking for an AI/ML engineer to design, build and run AI-powered applications on AWS. This includes large language model (LLM) applications, AI agents, retrieval-augmented generation (RAG), and classic machine learning. You’ll take ideas from prototype to production, with a focus on reliability, security, cost and measurable quality. The role is hired at two levels.
Responsibilities
- Design and build LLM applications: prompt design, RAG pipelines, tool or function calling, and multi-step agents.
- Build and deploy models and AI services on AWS: Amazon Bedrock, SageMaker, Lambda, API Gateway, ECS/EKS, Step Functions, DynamoDB, S3 and OpenSearch.
- Set up evaluation: offline test sets, LLM-as-judge, regression checks, and tracking of quality, latency and cost.
- Add observability and tracing for AI workloads, for example OpenTelemetry, CloudWatch, LangFuse or similar.
- Build in safety from the start: guardrails, prompt-injection defences, access control, human approval for risky actions, and handling of personal data.
- Write production-quality, tested code and maintain CI/CD and infrastructure as code (Terraform or CDK).
- Work with product, platform and security teams; write clear design documents and runbooks.
- Keep up with the field and assess new models, frameworks and AWS services.
Requirements
AI/ML Engineer (about 3-6 years)
- Strong Python; solid software engineering (testing, code review, APIs, Git).
- Hands-on experience shipping at least one LLM or ML system to production.
- Practical AWS experience: Bedrock and/or SageMaker, Lambda, IAM, VPC basics, and serverless or container deployment.
- Experience with RAG and vector search (OpenSearch, pgvector or similar) and embeddings.
- Experience with agent or orchestration frameworks (for example LangGraph, LangChain, LlamaIndex, Strands, Bedrock Agents) and protocols such as MCP.
- Understanding of evaluation methods, model limits, hallucination, and cost and latency trade-offs.
- Infrastructure as code (Terraform or CDK) and CI/CD.
- Security awareness: OAuth/OIDC, least-privilege IAM, secrets management.
- Good Python and ML fundamentals (supervised learning, model evaluation, basic statistics).
- A degree in CS, Data Science or a related field, or equivalent practical experience.
- Hands-on project work with LLM APIs, RAG or ML models (internships, open source or personal projects count).
- Familiarity with core AWS services; an AWS certification is a plus.
- Eagerness to learn production engineering practices: testing, monitoring and security.
Nice to have
- AWS certifications: Machine Learning Engineer - Associate, AI Practitioner, Solutions Architect. Fine-tuning or adapting models (LoRA, distillation) and MLOps tools (MLflow, SageMaker Pipelines).
- Experience with multi-agent systems, agent-to-agent (A2A) communication, or policy engines.
- Docker, Kubernetes (EKS) and Linux.
- Experience with responsible-AI practices and governance.
What we look for
- Clear thinking and plain communication with technical and non-technical colleagues.
- Ownership: builds, measures and improves things end to end.
- Judgement about when AI is, and isn’t, the right tool.
Qualifikationen
Job Description: AI/ML Engineer
About the role
We’re looking for an AI/ML engineer to design, build and run AI-powered applications on AWS. This includes large language model (LLM) applications, AI agents, retrieval-augmented generation (RAG), and classic machine learning. You’ll take ideas from prototype to production, with a focus on reliability, security, cost and measurable quality. The role is hired at two levels.
Responsibilities
- Design and build LLM applications: prompt design, RAG pipelines, tool or function calling, and multi-step agents.
- Build and deploy models and AI services on AWS: Amazon Bedrock, SageMaker, Lambda, API Gateway, ECS/EKS, Step Functions, DynamoDB, S3 and OpenSearch.
- Set up evaluation: offline test sets, LLM-as-judge, regression checks, and tracking of quality, latency and cost.
- Add observability and tracing for AI workloads, for example OpenTelemetry, CloudWatch, LangFuse or similar.
- Build in safety from the start: guardrails, prompt-injection defences, access control, human approval for risky actions, and handling of personal data.
- Write production-quality, tested code and maintain CI/CD and infrastructure as code (Terraform or CDK).
- Work with product, platform and security teams; write clear design documents and runbooks.
- Keep up with the field and assess new models, frameworks and AWS services.
Requirements
AI/ML Engineer (about 3-6 years)
- Strong Python; solid software engineering (testing, code review, APIs, Git).
- Hands-on experience shipping at least one LLM or ML system to production.
- Practical AWS experience: Bedrock and/or SageMaker, Lambda, IAM, VPC basics, and serverless or container deployment.
- Experience with RAG and vector search (OpenSearch, pgvector or similar) and embeddings.
- Experience with agent or orchestration frameworks (for example LangGraph, LangChain, LlamaIndex, Strands, Bedrock Agents) and protocols such as MCP.
- Understanding of evaluation methods, model limits, hallucination, and cost and latency trade-offs.
- Infrastructure as code (Terraform or CDK) and CI/CD.
- Security awareness: OAuth/OIDC, least-privilege IAM, secrets management.
- Good Python and ML fundamentals (supervised learning, model evaluation, basic statistics).
- A degree in CS, Data Science or a related field, or equivalent practical experience.
- Hands-on project work with LLM APIs, RAG or ML models (internships, open source or personal projects count).
- Familiarity with core AWS services; an AWS certification is a plus.
- Eagerness to learn production engineering practices: testing, monitoring and security.
Nice to have
- AWS certifications: Machine Learning Engineer - Associate, AI Practitioner, Solutions Architect. Fine-tuning or adapting models (LoRA, distillation) and MLOps tools (MLflow, SageMaker Pipelines).
- Experience with multi-agent systems, agent-to-agent (A2A) communication, or policy engines.
- Docker, Kubernetes (EKS) and Linux.
- Experience with responsible-AI practices and governance.
What we look for
- Clear thinking and plain communication with technical and non-technical colleagues.
- Ownership: builds, measures and improves things end to end.
- Judgement about when AI is, and isn’t, the right tool.

