Senior AI Solutions Engineer

About the role

As a Senior AI Solutions Engineer, you'll design, build, and run AI solutions that make a real difference in day-to-day decision-making across the business. This is a hands-on engineering role focused on shipping AI into production, not prototypes.

You'll work across Python + Java microservices, LLM/RAG systems, vector search, and data pipelines, deploying to AWS (incl. Bedrock) and Azure (Azure AI Foundry). You'll partner closely with the Lead AI Solutions Engineer, platform engineers, analysts, and data teams to deliver scalable capabilities that are secure, observable, and maintainable.

What you'll do

Build production AI systems (LLMs + RAG)

• Design and implement RAG-powered services (assistants, chat experiences, semantic search) using modern LLM patterns

• Improve retrieval quality through embeddings, metadata enrichment, ranking strategies, and evaluation feedback loops

• Build modular components that can be reused across multiple use cases and domains

Develop scalable APIs and microservices

• Build and maintain backend services and APIs using Python (FastAPI/LangChain/Hugging Face) and Java

• Create clean service boundaries, versioned APIs, and secure integration patterns for enterprise environments

• Produce high-quality documentation and maintain an engineering standard that scales beyond one team

Engineer reliable data and embedding pipelines

• Build and operate pipelines for ingestion, embedding generation, chunking strategies, and metadata processing

• Orchestrate ETL/ELT workflows using Airflow for batch and near-real-time use cases

• Ensure governance, security, and privacy requirements are met (and provable)

Operate in cloud with strong engineering hygiene

• Deploy solutions across AWS and Azure, using CI/CD and IaC to keep releases safe and repeatable

• Containerise and run workloads with Docker and Kubernetes, working with Platform Engineering on Kindred Cloud

• Build with production realities in mind: logging, monitoring, failure handling, scalability, and cost controls

Own semantic search and vector database performance

• Implement and optimise vector search using PGVector / ChromaDB, including indexing strategies and query performance

• Work with Sentence Transformers / OpenAI embeddings and similarity techniques (e.g., cosine similarity) to improve precision/recall

Collaborate, influence, and raise the bar

• Work across teams to align on design choices, integration patterns, and shared reusable components

• Mentor others through reviews, pairing, and knowledge-sharing sessions

• Bring pragmatic innovation: test new approaches, keep what works, and productise it

What success looks like

• AI features move from idea → production with measurable adoption and value

• RAG systems deliver relevant, trustworthy outputs with clear performance indicators

• Services are secure, observable, and operationally stable (not fragile demos)

• Engineers and stakeholders trust the platform and can build on it without reinventing the wheel

What we're looking for

• 5+ years in backend engineering, data engineering, or AI/ML integration roles

• Strong hands-on skills in Python and solid experience with Java (or deep JVM ecosystem experience)

• Practical experience building with LLMs, embeddings, semantic search, and RAG-style architectures

• Experience with vector databases (PGVector/ChromaDB or similar) and retrieval optimisation

• Strong delivery habits: CI/CD, Docker, Kubernetes, and Infrastructure as Code (Terraform/CloudFormation)

• Cloud experience across AWS (EC2, S3, Lambda, Bedrock, CodePipeline etc.) and/or Azure AI Foundry

• Comfortable working with stakeholders, ambiguity, and trade-offs — you can turn fuzzy problems into shipped outcomes

Nice to have

• Experience fine-tuning or adapting models for domain use cases

• Experience building internal developer platforms or reusable AI components

• Experience with evaluation/observability for GenAI systems (quality, latency, cost, drift, safety)

• Prior experience in regulated environments or with identity/security integrations (SSO, IAM)

Why join

• Work on real AI products used across the organisation — not proofs of concept

• Modern stack: LLMs, RAG, vector search, Kubernetes, multi-cloud

• Strong cross-functional collaboration with data, platform, and product teams

• Space to innovate, improve foundations, and build capabilities that scale