Job Category: IT
Type: employment contract (CIM)
Location: Remote from Romania
Level: Senior
Status: Open
Requirements: Azure CI/CD Kubernetes Python (Programming Language) Retrieval-Augmented Generation (RAG) AI infrastructure LLM systems
This role is at our client, a global SaaS technology provider specialising in governance, compliance, entity management, and operational software for financial services organisations. They develop cloud-based solutions that help businesses automate workflows, manage regulatory obligations, streamline corporate administration, and improve data governance across multiple jurisdictions. It serves international clients in highly regulated industries with a focus on digital transformation, operational efficiency, and compliance automation.
The Role
Responsibilities
- Design and maintain ingestion, chunking, embedding, and indexing pipelines for LLM training and RAG datasets using LangChain or equivalent frameworks.
- Manage lifecycle of prompt templates, embedding models, LLM chains, and evaluators; support benchmarking of foundation models and retrieval strategies.
- Build and operate AI infrastructure (Azure AI Foundry, Azure OpenAI, Azure App Services) including secure hosting for RAG apps, vector databases, and agent runtimes.
- Define CI/CD pipelines for LLM artefacts (datasets, prompts, model configs, eval suites) via Azure DevOps, in collaboration with DevOps engineers.
- Establish observability for LLM systems (telemetry, latency, cost, hallucination rates, retrieval quality) using LangSmith or equivalent; integrate automated evaluation and quality gates into CI/CD.
- Apply responsible AI, privacy, access control, and logging standards; support red-team testing for prompt injection and model risks.
- Partner with AI developers, Portfolio Architects, and Product Owners to integrate LLM/RAG components into product features and shape technical roadmap.
- Stay current on LLMOps, RAG optimization, and vector search trends; help establish early LLMOps best practices as the org scales.
Must-haves
- 4+ years in software/data/ML engineering or DevOps, preferably cloud-based.
- Strong Python; hands-on with LangChain, Semantic Kernel, or equivalent AI frameworks.
- Experience operating Azure AI Foundry, Azure OpenAI, Azure App Services, Azure Storage.
- Familiarity with vector databases and retrieval pipelines (Azure AI Search, Pinecone, Chroma, Redis Vector).
- Strong CI/CD, version control, and environment management skills (Azure DevOps preferred).
- Experience with Kubernetes (AKS) and containerized deployments.
- Experience with observability tooling (Azure Monitor, logging, tracing, metrics).
- Bonus: front-end/service dev skills (React, TypeScript, C#).
- Bachelor’s in CS/Software/Data Engineering or equivalent experience; strong problem-solving and cross-team communication skills.
Nice-to-haves
- Front-end/service dev skills (React, TypeScript, C#).
Interested in finding out more?
To get a better idea of the project you could be a part of, start a discussion with our EDUROM Recruitment consultant: below you can send us your contact information and relevant experience for this role.