Senior AI Engineer

Bayer CropScience Limited; Warszawa; On-site; PLN 100,000 - 140,000; Full time

Job summary

Bayer CropScience Limited is looking for an experienced professional to industrialize and scale AI prototypes into secure IT products. The role involves designing, implementing, and operating cloud-native APIs using Python and FastAPI, and developing protocols to safely expose enterprise tools.

The ideal candidate has extensive experience in AI/ML engineering, strong cloud expertise with AWS and Azure, and excellent communication skills in English. A Master's degree in Computer Science or a related field is required.

Qualifications

• 5+ years of professional experience in AI/software/ML engineering.
• Proficiency in AWS and/or Azure services.
• Strong software engineering fundamentals including testing and error handling.

Responsibilities

• Industrialize and scale GenAI prototypes into IT products.
• Design and operate cloud-native APIs for AI workloads.
• Implement CI/CD pipelines with automated testing.

Skills

Advanced Python

API Development (REST/gRPC)

CI/CD (GitHub Actions)

Docker

AWS expertise

Kubernetes experience

Infrastructure as Code (Terraform)

LLMs and embeddings

Monitoring and traceability (OpenTelemetry)

Strong problem-solving skills

Education

Master's degree in Computer Science or related field

Tools

AWS (Lambda, ECS/Fargate/EKS)

Azure Functions

PostgreSQL

Databricks

Job description

Responsibilities

Industrialize and scale successful GenAI prototypes into secure, resilient IT products for Enabling Functions.

Design, implement, and operate cloud-native APIs and microservices for AI workloads using Python and FastAPI, following schema‑first design (OpenAPI/gRPC).

Develop Model Context Protocol (MCP) servers (FastMCP) to safely expose enterprise tools and data to agents, ensuring robust permissions and auditing.

Architect agent workflows with LangChain, LangGraph, and PydanticAI (tool calling, memory, event‑driven orchestration).

Build reliable text‑to‑sql solutions and/or RAG services with high‑quality embeddings, indexing, reranking, and caching for performance and cost efficiency.

Implement CI/CD pipelines (GitHub Actions) with automated testing.

Deploy on AWS and/or Azure (containers, serverless, API gateways, managed databases, object storage, secrets).

Ensure end‑to‑end observability: structured prompt/response logging with redaction, token/latency/cost tracking, OpenTelemetry tracing, and model/agent monitoring (e.g., Langfuse/LangSmith/MLflow).

Establish safety and quality controls: evaluation pipelines, prompt/chain regression tests, content guardrails, and injection defenses.

Collaborate across Data Science, MLOps/DevOps, Architecture, Product, and Business to align solutions with outcomes; contribute to stack decisions and cost/scalability trade‑offs.

Promote continuous learning via code reviews, tech talks, and mentoring on AI engineering best practices.

Qualifications

Master's degree (or equivalent) in Computer Science, Data/AI, Mathematics, or a related field; PhD is an advantage.

5+ years of professional experience in AI/software/ML engineering, with end‑to‑end product delivery in production environments.

Advanced Python and production‑grade API development (REST/gRPC), authN/authZ (OAuth2/OIDC), rate limiting.

Containerization (Docker) expertise; Kubernetes experience is a plus.

Proficiency in AWS and/or Azure: AWS (Lambda, ECS/Fargate/EKS, API Gateway, S3, RDS, Secrets Manager, Bedrock); Azure (Functions, AKS, API Management, Storage, PostgreSQL/Cosmos DB, Key Vault, Azure OpenAI).

Strong CI/CD knowledge, especially GitHub Actions.

Infrastructure as Code (Terraform preferred).

Solid grasp of LLMs and embeddings: context management, tool calling, streaming, and latency/cost trade‑offs.

Monitoring and traceability mindset: OpenTelemetry, Langfuse/LangSmith.

Strong software engineering fundamentals: testing, code reviews, error handling, reliability/resilience.

Excellent problem‑solving and communication skills; fluent in English (written & spoken).

Preferred

• Hands‑on with agent frameworks (LangChain, LangGraph, PydanticAI) as well as FastAPI and FastMCP for MCP server development.
• RAG and vector search proficiency (pgvector, OpenSearch).
• Experience with relational databases (e.g., PostgreSQL); Databricks experience a plus.