SDC AI Developer - Senior

Job description

Requisition ID:

1719828

At EY, we’re all in to shape your future with confidence.

We’ll help you succeed in a globally connected powerhouse of diverse teams and take your career wherever you want it to go. Join EY and help to build a better working world.

SDC AI Developer - Senior

The Opportunity

Supports the delivery of solution or infrastructure development services for AI/ML initiatives, applying strong technical capability and hands-on engineering experience. Contributes to the design, development, delivery, and maintenance of AI-enabled solutions or infrastructure while aligning to relevant engineering standards and project delivery expectations. Understands user requirements and helps translate them into sound technical designs and implementation plans. Contributes to the integration of AI/ML capabilities into broader enterprise solutions, with a focus on quality, scalability, and user impact.

Your key responsibilities

• Develop, test, deploy, and support production-grade AI/ML, generative AI, and intelligent automation solutions.
• Solve complex technical problems across development, integration and production support through coding, debugging, testing, troubleshooting, and structured design remediation.
• Translate user requirements into technical designs, APIs, workflows, and supportable implementation patterns.
• Build and integrate LLM, RAG, and agentic solution components into enterprise solutions, applications and platforms.
• Support project delivery through disciplined execution, estimation, documentation, status’ communication, and risk identification.
• Participate in design reviews, providing thoughtful trade-off analysis and implementation input.
• Use modern AI-assisted software engineering tools such as Claude Code, Codex, or equivalent agentic coding platforms as part of day-to-day engineering delivery to improve delivery speed, code quality and engineering efficiency.

AI and Engineering Skills:

Gen AI Foundational:

• Experience designing, building, and maintaining production-grade LLM applications, including end-to-end pipelines from data ingestion through model output delivery (e.g. Azure OpenAI, AWS Bedrock, Google Vertex AI etc.).
• Demonstrated practical experience building retrieval-augmented systems that ground model outputs in enterprise knowledge sources, including chunking strategies, embedding pipelines, and retrieval optimization (e.g. LlamaIndex, LangChain, Pinecone, Weaviate, Azure AI Search, pgvector etc.).
• Working technical knowledge of embedding models, vector search, and semantic retrieval patterns used to ground LLM outputs in enterprise knowledge sources (e.g. OpenAI Embeddings, Azure AI Search, pgvector etc.).
• Proficiency in prompt engineering techniques including zero-shot, few-shot, chain-of-thought, and structured output design, with the ability to systematically evaluate and iterate on prompt performance (e.g. DSPy, PromptFlow etc.).

Agentic and LLM Ops:

• Experience designing and building agentic systems including multi-agent orchestration patterns, tool use, and memory design across single and multi-step workflows (e.g. LangGraph, AutoGen, CrewAI, Semantic Kernel, NVIDIA NIM etc.).
• Ability to debug, troubleshoot, and remediate production LLM and agentic systems including failure diagnosis across retrieval, orchestration, and generation layers.

Software Engineering:

• Hands-on software engineering proficiency in Python, with the ability to write clean, modular, production-quality code for LLM pipelines and agentic applications.
• Experience working with structured and unstructured data sets to support LLM application development, including data curation, preparation, and quality validation for model inputs and model responses.
• Working familiarity with RESTful and event-driven API patterns including asynchronous workflows, service boundaries, and integration of enterprise data sources to expose LLM and agentic capabilities.
• Practical understanding of containerization and orchestration concepts for packaging and deploying LLM applications in cloud environments (e.g. Docker, Kubernetes, Azure Container Apps, AWS ECS etc.).
• Understanding of software engineering best practices as applied to ML systems, including modular code design, testing patterns for AI pipelines, and data quality validation.
• Familiarity with Data Monitoring and Data Observability in cloud environments (Open Telemetry, Azure Application Insights etc.).
• Exposure to CI/CD and operationalization practices for AI systems, including model and workflow deployment, versioning, environment promotion, and release support in cloud or containerized environments.

To qualify for the role you must have

• A bachelor's or master’s degree
• Minimum of 2 years of related work experience applied engineering experience, including meaningful experience in AI/ML engineering roles
• Clear communicator able to explain complex AI system behavior and trade‑offs to technical and non‑technical stakeholders, including risk and compliance.
• Strong ownership and accountability, taking responsibility for AI systems from design through production and issue resolution.
• Able to operate effectively as requirements, regulations, and technologies evolve.
• Collaborative and cross‑functional, working closely with engineering, product, teams.

Ideally, you’ll also have

• Partner with Development, Engineering, Product, Data, Architecture, and project leadership teams to deliver high-value AI capabilities.
• Ability to build and maintain model observability pipelines including tracing of multi-step agentic reasoning chains, output degradation detection, and behavioral drift monitoring in production (e.g. LangSmith, Arize, Datadog, Azure Monitor etc.).
• Familiarity with LLM fine-tuning approaches including instruction tuning and preference optimization, with an understanding of when fine-tuning is appropriate versus prompt-based solutions (e.g. LoRA, QLoRA, PEFT, NeMo Framework etc.).
• Familiarity with responsible AI principles including bias and fairness evaluation, human-in-the-loop design, and explainability approaches in the financial services contexts.
• Familiarity with data pipeline design for AI workloads including ingestion, transformation, and quality validation.
• Familiarity with cloud-based platforms for building, training, and deployin