Applied AI Site Reliability Engineer I (PxE A&A)

About the Role

As an Applied AI Site Reliability Engineer I, you will actively engage in your engineering craft, taking a hands-on approach to the reliability, performance, and operational integrity of high-visibility products and platforms and the environments they run in. Your expertise will be pivotal in keeping production safe, performant, and cost-effective, while driving tangible value for Deloitte's engineering investments. You will leverage your engineering craftsmanship across cloud platform engineering, observability, and performance and reliability engineering—together with applied AI fluency that lets you reliably operate AI and agentic workloads alongside the rest of the portfolio—consistently demonstrating your strong track record in operating high-quality, resilient systems at scale. The ideal candidate will be a dependable team player, collaborating with cross-functional teams to uphold production standards, safeguard environments, and support the admission of systems into production with confidence.

Key Responsibilities

• Outcome-Driven Accountability: Embrace and drive a culture of accountability for reliability, performance, and cost outcomes, measured in service-level objectives and error budgets, not raw uptime. Operate the products, platforms, and environments you support to meet their SLOs within budget, and track incident trends and toil to prioritize the work that most improves reliability—ensuring high-quality, lean operational designs that keep production safe and resilient.
• Technical Leadership and Advocacy: Serve as the technical advocate for production reliability and operability, ensuring systems are admissible, performant, safe to run, and able to degrade gracefully when failure occurs. Uphold production standards, participate in the design of observability, performance and resilience testing, and operational tooling, and support the admission of systems into production—gating release on error budgets and automated reliability checks, and owning the readiness verification, environment integrity, and operational support that follow.
• Engineering Craftsmanship: Maintain accountability for the operational integrity of production and pre-production environments, and for the production standards that systems are admitted against. Own SLOs and error budgets; build and operate production observability—codified, version-controlled dashboards and SLO-driven, actionable alerting that detects before impact, plus the feedback loop into engineering; run performance, ambient-noise, and chaos testing to verify readiness; and guard environments against drift. Stay hands-on, self-driven, and continuously learn new approaches, languages, and frameworks—operating as an infrastructure-focused engineer, not a tool operator. Create technical specifications, runbooks, and shared playbooks; lead blameless postmortems that turn incidents into learning and systemic fixes; write high-quality, supportable automation to ensure all reliability KPIs (availability, performance, and cost) are met or exceeded. Demonstrate collaborative skills to work effectively with diverse teams.
• Customer-Centric Engineering: Develop lean operational solutions through rapid, inexpensive experimentation to meet the reliability needs of the engineering teams and the business. Engage with those teams before, during, and after delivery, co-defining service-level objectives and operational readiness so the right safeguards are in place at the right time, without becoming a bottleneck to delivery.
• Incremental and Iterative Delivery: Adopt a mindset that favors action and evidence over extensive planning. Utilize a leaning-forward approach to navigate complexity and uncertainty, hardening reliability through incremental, measurable improvements—progressive resilience testing and SLO refinement—rather than big-bang interventions, and keeping operations supportable and maintainable.
• Cross-Functional Collaboration and Integration: Work collaboratively with empowered, cross-functional partners: engineering, platform engineering, security and risk, data governance, and engineering leadership and architecture. Uphold production standards and integrate their constraints so that the reliable, performant, and compliant path is the operative path. Co-define service-level objectives with the teams you support, verify readiness, and support the admission decision into production—holding the segregation-of-duties line as a dedicated, embedded function while partnering with security and risk on the control objectives you enforce. Foster a collaborative environment that enhances team synergy and innovation.
• Advanced Technical Proficiency: Develop expertise in site reliability and modern production engineering—cloud platform ownership, observability (metrics, tracing, logging), performance and capacity engineering, chaos engineering, and cloud/AI cost engineering—together with applied AI fluency to operate AI and agentic workloads reliably, including AI and Agentic SSDLC, delivering production operations with full automation from discovery to production to operations and all quality checks through the SSDLC lifecycle. Learn to be a role model, leveraging these techniques to optimize reliability, performance, and operational delivery. Demonstrate understanding of the full lifecycle of platform and product development, focusing on continuous improvement and learning.
• Domain Expertise: Quickly acquire domain knowledge of the products and platforms you operate—and, where they are AI-infused, their distinct production failure modes such as drift, train/serve skew, latency and output variance, and token/GPU cost anomalies. Translate reliability needs, reference architectures, and operational requirements into service-level objectives, runbooks, and production tooling. Be a valuable, flexible, and dedicated team member, supportive of teammates, and focused on quality and tech debt payoff.
• Effective Communication and Influence: Exhibit exceptional communication skills, capable of articulating complex technical concepts clearly and compellingly. Inspire and influence teammates and product teams through well-structured arguments and trade-offs supported by evidence. Create coherent narratives that align technical solutions with business objectives.
• Engagement and Collaborative Co-Creation: Engage and collaborate with product engineering teams at all organizational levels, including customers as needed. Build and maintain constructive relationships, fostering a culture of co-creation and shared momentum towards achieving product goals. Align diverse perspectives and drive consensus to create feasible solutions.
• Strategic Vision and Alignment: Craft and articulate a vision for the Applied AI engineering platform—the frameworks, tooling, accelerators, control plane, and data platform that every other engineering team builds on—in alignment with the Business Strategy and US Deloitte Technology strategy. Treat the platform as a product, mapping the needs of consuming engineering teams to reusable, self-service capabilities. Collaborate with diverse stakeholders across product, engineering, SRE, security and risk, data governance, and infrastructure at all organizational levels.
• Advocacy and Technology Roadmap: Advocate for, develop, and communicate the integrated platform engineering strategy and its implementation roadmap, positioning the platform as a product to the engineering teams who consume it and to business stakeholders. Ensure the organization is well-informed about objectives, KPIs, adoption, maturity, compliance, and progress. Promote a culture of reuse, quality, and speed—solving recurring problems once, well, for many teams to reduce per-team build and operate toil, maximize engineering leverage, and minimize total cost of ownership.
• Craft Mastery and Objectives Realization: Define, measure, and drive the achievement of platform KPIs and NFRs—adoption, reliability-by-design, developer experience, performance, scalability, and security. Establish and evolve platform reference architectures, golden paths, and reusable components—including the AI control plane (with the model/LLM gateway) and agent-runtime patterns, MLOps/LLMOps tooling, model serving, retrieval and vector infrastructure and RAG-as-a-Service, Eval-as-a-Service and observability tooling, prompt management, and feature store; governance-as-code enforcement, control codification, and ontology management; the data platform (data catalog and marketplace, DataOps, data management and enterprise integration, master data, pipelines, and metadata/lineage); and CI/CD at platform scale—delivered through the AI and Agentic SSDLC as self-service, governed capabilities with full automation from discovery to production to operations and quality checks through the SSDLC lifecycle. Remain hands-on with design, architecture, and code—contributing to the leverage and velocity of the engineering teams and product groups that build on the platform—while reviewing code, driving tech-debt reduction, and experimenting with new technology.
• Capability Evolution and Development: As a recognized engineering leader, mentor and develop engineers and emerging engineering leaders, coaching modern platform engineering practices—internal developer platforms, distributed systems, the AI control plane and agentic patterns, MLOps/LLMOps, cloud-native design and container orchestration, platform-scale infrastructure-as-code, and the data platform. Lead by example through thought leadership—showcasing experiments internally, speaking at conferences, publishing whitepapers or blogs, and leading R&D collaborations, including with academia. Cultivate a growth mindset and modern engineering behaviors across the organization.
• Iterative Value Delivery: Embrace an iterative and incremental approach to platform engineering, delivering platform capabilities as lean, adoption-validated increments rather than big-bang builds. Apply a leaning-forward approach and empirical methods to navigate complexity and uncertainty, keeping capabilities supportable and maintainable and ensuring each increment delivers value and stays aligned with consuming-team and business goals.
• Customer-Centric Problem Solving: Maintain a relentless focus on the real needs of the engineering teams who consume the platform—your primary customers—developing lean capabilities through rapid, inexpensive experimentation and engaging those teams before, during, and after delivery so the right capability is delivered at the right time and adopted, not shelved. Minimize unnecessary technical complexity and avoid overengineering, and drive toward peak performance through continuous learning and collaborative execution.
• Expert Proficiency and Continuous Improvement: Possess deep expertise in modern platform engineering and AI/ML and data infrastructure, with a keen ability to identify inefficiencies and opportunities for innovation across the platform and the engineering operating model. Continuously enhance the operating model to be lean, adaptable, and responsive—guiding and transforming the organization to embrace lean principles and foster a culture of innovation.
• Tech/Quality Risk Management: Establish and evolve the enterprise tech-stack conformance baseline, platform standards, and quality benchmarks that engineering teams build against—embedding consistency, compliance, and security by default through guardrails (PII redaction, prompt-injection and content controls, tool permissioning and RBAC) and multi-tenant isolation, so that the secure, compliant, and reliable path is the easy path. Build capabilities that are operable, instrumented, performant, and drift-resistant by design to the production standards set by SRE, delivering them admissible-by-design and co-defining service-level objectives with SRE, who own the admission decision. Proactively identify technical risks and develop mitigation strategies through proactive problem-solving and contingency planning.
• Influential Communication: Influence, persuade, and drive decision-making across the organization. Communicate effectively in both written and verbal forms, crafting clear, structured arguments and technical trade-offs supported by evidence.
• Organizational Engagement and Collaboration: Engage stakeholders at all levels—from team members to middle management to executives—and collaborate with empowered, cross-functional partners across engineering, SRE, security and risk, and data governance, integrating their constraints into the paved roads. Build collaborative, constructive relationships and co-create momentum and value across the organization.

Qualifications

• A bachelor's degree in computer science, software engineering, data science, machine learning, or related discipline.
• Candidates must be located within a commutable distance to one of the select locations available for this role.
• Ability to work in your local office at a minimum of 3 days per week.
• 10+ years of software and platform engineering experience with most of the following: Angular, React, NodeJS, Python (Mandatory), C#, .NET, Java, Rust, SQL/NoSQL, REST/SOAP/GraphQL, SSO/MFA, PyTorch, TensorFlow, LangChain, LangGraph, LangSmith, LangFuse, Terraform, as well as unit, integration, and end-to-end testing frameworks & tools, specifically BDD, Gherkin, Cucumber, Playwright, and Selenium.
• 7+ years of experience architecting and delivering enterprise solutions and platforms on modern technology stacks (e.g., API Gateways, Message Brokers, Queuing Services, Workflow Automation & Orchestration, ETL/ELT, Event Streaming, Real-Time Data Processing, Service Mesh) and cloud-native engineering, using FaaS, PaaS, and micro-services on any of the cloud hyperscalers such as Azure, AWS, or GCP, including leveraging their AI/ML services such as Azure OpenAI, AWS Bedrock, or Vertex AI, as well as container orchestration (Kubernetes, Docker), Big Data, data platform tooling such as Informatica, Databricks, or Delphix (spanning data governance and observability, data fabric, and data anonymization), CI/CD and infrastructure-as-code at platform scale, and distributed systems.
• 5+ years of experience designing, building, and operating AI/ML platform or infrastructure, with hands-on experience across building tooling for MLOps/LLMOps, model serving, retrieval and vector infrastructure, and eval/observability instrumentation for LLM integration (OpenAI, Anthropic, or open-source models).
• 2+ years of experience establishing engineering platform standards and golden paths, including actively leading, mentoring, and guiding team members in the adoption and continuous improvement of these standards, treating the platform as a product.
• Prior experience with AI control-plane and agent-runtime patterns: model/LLM gateway, A2A and MCP integration, agent runtimes (e.g., Google ADK, Amazon Bedrock AgentCore), guardrails (PII redaction, prompt-injection, content, tool permissioning/tool-RBAC), policy-as-code, and multi-tenant isolation.
• Prior experience with enterprise data platform engineering: data pipelines, self-service and data-product enablement, governance-as-code enforcement, and metadata/lineage.
• Prior software engineering experience with the understanding of Business Context Diagrams (BCD), sequence/activity/state/entity relationship/data flow diagrams, OOP/OOD, data structures, algorithms, and code instrumentations, and AI-augmented spec-driven development.
• Prior experience using methodologies & tools such as XP, Lean, DevSecOps, SRE, ADO, GitHub, SonarQube, and ArgoCD to deliver high-quality platforms and products rapidly.
• Ability to travel 10%, on average, based on the work you do and products you build.