AI Engineer - EY GDS

Job Description: AI & Data – AI Engineer

Location: Buenos Aires - Argentina (Hybrid)

Clients: US‑based Enterprise Clients

About the Role

The Senior AI Engineer designs, builds, and ships enterprise-grade AI/ML and LLM-based solutions. This role focuses on hands-on engineering, high-quality delivery, and strong collaboration with cross-functional teams.

Key Responsibilities

Design, build, and deploy AI/ML and LLM-based solutions in enterprise environments.

Collaborate with cross-functional teams (Data Engineering, Cloud, Product) to deliver scalable AI systems.

Ensure high engineering standards, maintainability, and best practices.

Participate in code reviews, architecture discussions, and solution design.

Support continuous improvement of AI delivery processes and tooling.

Skills & Qualifications

• Python & Development

• Advanced Python (3–6 years);

• FastAPI;

• scikit-learn;

• API design;

• clean code;

• Preferred: intermediate SQL, Design patterns (clean architecture/hexagonal); microservices; advanced testing; Docker

• What we evaluate: Code quality; API design; troubleshooting; software architecture discipline; applied SQL

• LLMs, RAG & Agents:

• End-to-end RAG; LangChain/LangGraph;

• Vector search (FAISS or similar);

• Fine-tuning (LoRA/QLoRA);

• Advanced evaluation (RAGAS/TruLens/DeepEval);

• Agent design

• Autogen;

• Preferred: Llama Index; custom retrievers

What we evaluate: Hallucination mitigation; grounding; cost/latency trade-offs; quality

• Cloud (Azure or Databricks):

• Cloud (Azure): Azure OpenAI; Azure AI Search; Azure ML; service integration; AKS/Container Apps; API Management

• Databricks: Advanced MLflow (registry/tracking/serving); Delta Lake; Unity Catalog; Feature Store; Vector Search

• Preferred: Workflows/DLT,

• What we evaluate: Secure & scalable architectures; integration; resilience, Pipelines; governance (Unity Catalog); productivity

• MLOps & Delivery:

• CI/CD (GitHub Actions/Azure DevOps);

• Docker;

• AKS/Kubernetes;

• End-to-end ML pipelines;

• Basic monitoring (latency, cost, failures)

• Preferred: AI observability (tracing/telemetry); advanced Bicep/Terraform

What we evaluate: Reliability; diagnostics; automation

• ML Fundamentals:

• Classic models;

• Advanced metrics & trade-offs;

• When to use classic ML vs. LLMs

• Preferred: Advanced/ensemble models

What we evaluate: Technical judgment; model validation

• Communication and other requirements:

• English: Fluent B2+ technical communication

• Autonomy in English, Technical clarity;

• Proactive

• Good at managing request gathering and handling

• Proactive communication