Lead Agentic AI Engineer – VP (Mississauga)

Citi•Mississauga, ON

27d•CA$120,800 - CA$170,800•Onsite

About The Position

Citi's Wholesale Technology organization is seeking an exceptional, hands-on Lead Agentic AI Engineer (VP) to design, build, and deploy cutting-edge agentic AI solutions. This role combines deep technical leadership with architectural ownership — driving adoption of LLMs, agentic workflows, and generative AI platforms to improve efficiency, automation, and risk reduction across Citi's global banking operations. You will operate with an AI-first mindset, emphasizing rapid prototyping, MVP-driven development, and iterative delivery of production-grade AI capabilities.

Requirements

6–10 years of relevant experience in an AI/ML development role, Applications Development, or Systems Analysis, with a substantial and demonstrated focus on Python technologies.
Minimum 2+ years of professional experience in software development with a focus on AI, prompt engineering, machine learning, and/or agentic AI systems.
Proven track record as a lead developer for agentic flow design, prompt design, and testing of autonomous AI systems with deep expertise in Google ADK.
Subject Matter Expert (SME) in at least one area of Applications Development, particularly Python application development (Django, Flask, FastAPI).
Programming Python (expert-level): FastAPI, Django, Flask, asyncio, PySpark — strong fundamentals in algorithms, data structures, concurrency, and design patterns.
Proficient in Java (Spring Boot, Spring Cloud), JavaScript/TypeScript (React, Next.js, Node.js), and SQL/data modeling.
Experience across AWS, Azure, and GCP with Docker, Kubernetes, and CI/CD pipelines.
Proficient in MLOps practices including model versioning, deployment, and lifecycle management
Strong foundation in secure API design, microservices, event-driven architecture, and distributed systems with expertise in testing, Git workflows, and performance optimization.
Deep expertise in LLMs (OpenAI GPT, Gemini, Claude, Llama) with hands-on experience in LangChain, LangGraph, LlamaIndex, AutoGen, CrewAI, and Google ADK.
Familiar with Vertex AI, MCPs, agent communication standards, and AI coding tools including GitHub Copilot, Devin, and Claude Code.
Proven experience building advanced RAG systems (hybrid search, re-ranking, metadata filtering) with vector databases including Pinecone, Weaviate, FAISS, pgvector, and ChromaDB.
Hands-on experience in PyTorch, TensorFlow, Keras, and Scikit-learn including fine-tuning and embeddings.

Nice To Haves

Performance Optimization: Redis, Hazelcast; low-latency distributed systems.
Data Engineering: ETL/ELT pipelines; Apache Spark, Kafka.
Frontend: React, Angular, Vue.js for full-stack capabilities.
Master’s degree preferred

Responsibilities

Lead end-to-end design, development, and deployment of large-scale agentic AI solutions using Google Agent Development Kit (ADK) and frameworks such as LangChain, LangGraph
Architect advanced multi-agent systems (perception, reasoning, planning, execution) integrating multiple LLM providers (OpenAI, Anthropic, Google Gemini).
Build AI-powered capabilities using Google Gemini, Vertex AI, Agent Development Kit (ADK), Google A2UI, vector databases, RAG pipelines, semantic search, and advanced prompt and context management.
Engineer autonomous agents incorporating planning, tool usage, memory management, and multi-step reasoning patterns.
Develop scalable, high-performance backend services in Python (FastAPI, asyncio) with resilient APIs, event-driven designs, and microservices architectures.
Build and maintain robust data pipelines working with SQL (Oracle, PostgreSQL) and NoSQL (MongoDB) databases.
Implement secure REST APIs and agent interfaces with strong authentication, authorization (OAuth), and encryption best practices.
Optimize AI agent performance, latency, and cost through prompt optimization, caching strategies, and vector index tuning.
Provide architectural guidance for Next-Generation AI (NGAI) initiatives, ensuring adherence to CTO guidelines and platform standards.
Develop and maintain a strategic roadmap for generative AI adoption, evaluating new models, techniques, and platforms.
Establish and govern best practices for the full AI development lifecycle: prompt engineering, model evaluation, MLOps, and data management.
Drive CI/CD practices integrating automated testing, agent evaluation, code quality gates, containerization, and cloud-native deployment pipelines.
Automate AI model quality, performance testing, and MLOps build processing in the CI/CD pipeline.
Mentor AI/ML Engineers on best practices in agentic AI development, Google ADK, and advanced AI technologies.
Champion MVP-driven delivery, rapid iteration, and A/B experimentation to achieve fast time-to-value.
Collaborate with business units to identify high-impact use cases and ensure AI solutions meet business goals.