Confirmed on the employer's own hiring board on Nov 16, 2026. First seen by Alion on Nov 16, 2026. Symphoni HR scores A on the Alion truth index.
The Applied AI Engineer will design and build core AI/ML solutions that power the enterprise AI capabilities of a large financial services organisation. The role owns significant AI subsystems, from agentic workflows and LLM applications to evaluation pipelines and MLOps tooling, delivering production-grade solutions that are secure, scalable and aligned with the architectural vision. The successful candidate bridges AI architecture and hands-on implementation, drives technical delivery and mentors junior engineers.
Our client is a leading global financial services organisation with a long-standing presence in banking and markets. They combine deep regulatory expertise with a strong focus on technology-led innovation, serving millions of customers worldwide. The culture emphasises integrity, responsible innovation and continuous learning across its technology teams.
Responsibilities
- Design, develop and deliver high-quality, scalable and performant AI software solutions using industry-aligned languages, frameworks and tools
- Build agentic AI systems and LLM applications, including multi-agent orchestration, tool use and multi-step AI workflows
- Develop RAG pipelines, prompt engineering strategies and model serving solutions using platforms such as vLLM
- Collaborate with product managers, designers and engineers to define requirements, devise solution strategies and ensure seamless integration
- Lead collaborative assignments, coach team members and participate in code reviews to promote code quality and knowledge sharing
- Implement secure coding practices, AI safety controls and unit testing to protect sensitive data and ensure reliability
- Build evaluation frameworks and observability for agentic systems in production
- Stay current with industry AI trends and contribute to internal technology communities
Requirements
- Bachelor's degree or above in Computer Science, Engineering, Mathematics or a related discipline
- 9+ years of software engineering experience, including at least 3 years building AI/ML systems, scalable ML infrastructure or model serving platforms
- Strong proficiency in Python (FastAPI, asyncio) and/or Go/Java, with production experience building APIs and microservices
- Hands-on experience with agentic frameworks (Strands, LangGraph, Google ADK), prompt engineering and RAG architectures
- Working knowledge of AWS services (EC2/EKS, S3, IAM, RDS, Bedrock/SageMaker), Docker, Kubernetes, CI/CD and Infrastructure-as-Code
- Solid understanding of distributed systems architecture, REST, gRPC and event-driven design, with observability tooling such as OpenTelemetry
- Ability to own system design from ideation to completion with limited supervision and communicate effectively with technical and non-technical stakeholders
- Desirable: financial services background, AI security frameworks (NIST AI RMF, OWASP LLM Top 10), fine-tuning (LoRA, RLHF), vector databases and Zero Trust identity systems
Skills
Python (FastAPI, asyncio) and/or Go/Java, Agentic AI systems and LLM applications, Prompt engineering, RAG architectures, agentic frameworks (Strands, LangGraph, Google ADK), GenAI/LLM systems and model serving (vLLM), Distributed systems architecture (microservices, REST, gRPC, event-driven), AWS services (EC2/EKS, S3, IAM, RDS, Bedrock/SageMaker), Docker and Kubernetes, CI/CD pipelines, Infrastructure-as-Code (CloudFormation, Terraform), observability (OpenTelemetry), Evaluation frameworks for agentic systems and observability, Vector databases, knowledge graphs, semantic search, memory systems, AI safety controls (guardrails, prompt injection detection, PII redaction), AI security frameworks (NIST AI RMF, MITRE ATLAS, OWASP LLM Top 10), Model fine-tuning (LoRA, RLHF, supervised fine-tuning), Responsible AI and model governance, Zero Trust architecture and identity systems (EntraID, OAuth2/OIDC), Financial services background and regulatory compliance, GPU workload management and inference optimisation (vLLM)

