6-Month contract – Inside IR35 – market rate
London based – hybrid working – up to 3 days a week onsite
Finance sector – must have previous experience
Role Overview
Senior AI / Data Engineer responsible for designing, building, and optimizing AI-driven data pipelines and integrations to enable a QAS-powered response suggestion capability embedded in Salesforce Service Cloud. The role focuses on scalable data processing, LLM integration, and continuous model improvement using production telemetry.
Responsibilities
- Design and implement Salesforce QAS integration architecture
- Build and optimize data pipelines supporting AI inference and feedback loops
- Integrate LLM capabilities (Amazon Bedrock) for response generation and embeddings
- telemetry data
- quality scoring
- usage analytics
- Work with structured and unstructured data sources:
- Microsoft Graph (SharePoint / Teams)
- Ensure data reliability, scalability, and performance
- Contribute to:
- Support:
- SIT/UAT phases
- production readiness
- hypercare and rollout to additional entities
Required Experience & Skills
Core
- 5–10 years of experience in Data Engineering / AI Engineering
- Strong experience in:
- Python / JVM-based backend development
- REST APIs / microservices
- Experience with cloud-native architectures on AWS
Data & AI
- Hands-on with:
- data pipelines (batch + streaming)
- embeddings / retrieval architectures
- Experience using:
- Snowflake (data platform integration, CDC concepts)
AWS Stack
- S3, RDS, SQS, EventBridge
- Containerized workloads (EKS/ECS)
- Strong understanding of:
- distributed systems
- performance optimization
- observability (e.g. Langfuse, logging/metrics)
Nice-to-Have
- Experience with:
- Salesforce Service Cloud integrations
- NLP / GenAI applications in customer service
- Exposure to:
- Amplitude or product analytics tools
- Knowledge of regulated environments (banking / capital markets)
Soft Skills
- Ability to work in cross-functional distributed teams
- Clear communication with business and technical stakeholders
#J-18808-Ljbffr…
