Senior AI/ML Engineer: RAG Pipelines

3 dni temu

Polska, Województwo śląskie Grape Up Pełny etat 180 000 zł - 260 000 zł Umowa

Automotive | Aviation | Manufacturing | Insurance

Seniority

Senior

Department

Software Engineering

AWS

Amazon Bedrock (Claude, Titan Embeddings, Knowledge Bases)

RAG Architecture & Prompt Engineering

Amazon SageMaker

Agent Frameworks (LangGraph/LangChain SDKs, MCP, AgentCore)

At Grape Up, we transform businesses by unlocking the potential of AI and data through innovative software solutions.

We partner with industry leaders in the automotive and aviation to build sophisticated Data & Analytics platforms that support production machine learning and AI use cases. Our solutions provide comprehensive capabilities spanning data storage, management, advanced analytics, machine learning, enabling enterprises to accelerate innovation and make trusted, data-driven decisions.

Responsibilities

  • Build and maintain RAG pipelines using Amazon Bedrock (Claude, Titan Embeddings, Knowledge Bases) to ground agent responses in real operational data
  • Design and implement agent reasoning and prompting strategies that support automated, near real-time decision-making
  • Apply MLOps practices on Amazon SageMaker – pipelines, versioning, evaluation – to keep models reliable as they move into production
  • Build multi-step, memory-aware agents using modern agent frameworks (e.g., LangGraph, MCP, AgentCore) with container-based deployment
  • Embed AI security practices – guardrails, PII redaction, IP protection – into every agent workflow, including any automated output that reaches an external party

Requirements

  • Master’s degree in computer science, Machine Learning, Data Engineering or a related field
  • 5+ years of overall software/ML engineering experience and 2+ hands-on GenAI/LLM/agentic experience
  • Strong, hands-on experience with AWS Cloud
  • Hands-on experience with Amazon Bedrock (including Claude models, Titan Embeddings, and Knowledge Bases)
  • Strong, current expertise in RAG architecture and prompt engineering (including OpenSearch vector search, Lambda-based ingestion pipelines, and metadata management alongside vectors)
  • Advanced experience with Amazon SageMaker, covering MLOps practices, pipelines, and model evaluation
  • Advanced, hands-on experience with agent frameworks such as LangGraph/LangChain SDKs or Strands Agents, including MCP and Amazon AgentCore in containerized environments (with working knowledge of long-term agent memory design)
  • Knowledge of AI security practices, including guardrails, PII redaction, and IP protection
  • Very good command of English and Polish, both written and spoken (B2+/C1) – daily collaboration within the team takes place in English

Nice to have

  • CI/CD skills for production GenAI and agentic workloads, with demonstrable, current hands-on knowledge of modern tooling
  • Experience with containerized ML workloads using Docker and Kubernetes
  • Experience with infrastructure-as-code (Terraform, CloudFormation, or similar)

Recruitment owner

Employee referral

Knowledge platforms

Equipment

Integration activities

Employee referral

Knowledge platforms

Equipment

Integration activities

Employee referral

Knowledge platforms

Equipment

Integration activities

Employee referral

Knowledge platforms

Equipment

Integration activities

Lunch & Learn

In-house Tech Up

Lunch & Learn

In-house Tech Up

Lunch & Learn

In-house Tech Up

Lunch & Learn

In-house Tech Up

Conferences & training

Feedback sessions

Business travel opportunities

Conferences & training

Feedback sessions

Business travel opportunities

Conferences & training

Feedback sessions

Business travel opportunities

Conferences & training

Feedback sessions

Business travel opportunities

Check out the next steps of your recruitment process

Application

Let’s talk about your experience and expectations, and what this role has in store for you.

Task

Put your skills to the test with a practical challenge.

Offer

We’re excited to welcome you onboard Let’s discuss the details.

Feedback

If you’re not lucky this time, we’ll tell you what you can improve to join us in the future.

Are you ready to become part of the Grape story?