Senior Lead AI Application Engineer

Job Description

As the Senior Lead AI Application Engineer, you will lead multi‑agentic application software engineering and full‑stack delivery of secure, cloud native microservices that embed pragmatic AI. You will build production UIs and APIs that orchestrate agent workflows end to end, integrate seamlessly with RBC’s CI/CD, observability, and security pipelines, and operate reliably in regulated environments. Leveraging LLMs, data pipelines, and GPU/multi cloud stacks, you will deliver resilient, scalable services that automate complex risk and security controls, accelerate regulatory readiness, and enable safe, auditable AI behaviors.

You will set standards, build reusable components, and mentor engineers across platforms and product teams.

• Design, build, and operate full‑stack agentic applications:
  React/Type Script front‑ends and secure back‑end services (Node.js/Express or Python/FastAPI or Java/Spring Boot) exposing REST/gRPC/GraphQL APIs; deliver UIs and APIs that orchestrate agent workflows end‑to‑end.
• Engineer multi‑agent systems (plan/act/reflect, tool use, delegation) with robust state and coordination across services; implement failure handling, retries, and idempotency using frameworks such as Lang Graph or Auto Gen.
• Deliver CI/CD and platform automation:
  Docker images, Git Hub Actions/Jenkins/Argo CD, IaC (Terraform/Helm), Kubernetes deployments, release/rollback and change‑control aligned to RBC standards.
• Implement context and data layers for agents: connectors (incl. MCP) to enterprise systems, streaming/eventing (Kafka), caching (Redis), vector search (pgvector/Milvus/Pinecone), and schema/PII controls to keep actions grounded and compliant.
• Build reusable SDKs, APIs, and templates for agent capabilities (inference, tools, RAG) with clear interfaces, code samples, and internal packages to accelerate teams.
• Establish runtime controls and observability:
  Open Telemetry traces/metrics/logs, Grafana/Prometheus dashboards, feature flags, rate limits/quotas, circuit breakers, and SLOs for latency, quality, and cost.
• Enforce Dev Sec Ops  and zero‑trust patterns: OAuth2/OIDC, mTLS, secrets (Vault/KMS), SAST/DAST, threat modeling; ensure privacy, data residency, and compliant deployments (NIST, SOX/ITGC) across AWS/Azure/on‑prem GPU.
• Partner across Security, Risk, and Platform; mentor engineers; influence reference architectures, standards, and documentation; drive design reviews for agentic services.

• Bachelors degree in Software Engineering, Computer Science or equivalent practical experience.
• 5+ years building production full‑stack systems:
  React/Type Script; back‑end with Node.js/Express, Java/Spring Boot, or Python/FastAPI; strong API design (REST/gRPC/GraphQL) and event‑driven patterns.
• Hands‑on multi‑agent engineering: inter‑agent communication, shared state, tool adapters, orchestration with Lang Graph/CrewAI/Auto Gen; RAG pipelines; vector DBs (pgvector/Milvus/Pinecone/FAISS); MCP‑based secure connectors.
• Cloud‑native delivery on Kubernetes with CI/CD (Docker, Git Hub Actions/Jenkins/Argo CD), automated testing (unit/integration/contract/E2E), and IaC (Terraform/Helm); experience with API gateways (Apigee/Kong) and service mesh (Istio/Linkerd).
• Data and integration: SQL (Snowflake/Postgres), No
  
  SQL (MongoDB), streams (Kafka), caching (Redis); integrating with enterprise systems (Service Now IRM/GRC) using resilient, secure patterns.
• Dev Sec Ops  and observability: OAuth2/OIDC, mTLS, Vault/KMS, SAST/DAST;
  Open Telemetry, Grafana/Prometheus; operating in regulated environments with NIST/SOX/ITGC controls.

• Advanced LLM operations: fine‑tuning (LoRA/PEFT), alignment (RLHF/DPO), optimization (Deep Speed), model serving (Triton/ONNX), and policy/guardrail engines. Distributed systems scale‑out:
  Ray/Sage Maker, back pressure and retry strategies, multi‑region/DR, and dual‑cloud/hybrid operations.
• Security and zero‑trust depth: service mesh/mTLS, WAF, secrets rotation, key management, and advanced SAST/DAST/IAST.
• Quantitative evaluation and product analytics: statistics for A/B tests and offline evals, human feedback loops,…