# Manager - Software Engineer (Agentic Engineering) - Tax Technology — Deloitte

- Location: Multiple Locations
- Posted pay: $138k–$262k/yr
- Posted: Jun 25, 2026
- Status: open (checked Nov 20, 2026)
- Apply (employer's site): https://apply.deloitte.com/en_US/careers/JobDetail/Manager-Software-Engineer-Agentic-Engineering-Tax-Technology/361415
- Employer record: https://visajobs.io/company/deloitte-tax.md

## Sponsorship history
- Tier: Good history
- Occupation (SOC): Project Management Specialists (13-1082)
- …in the whole US: 19
- With this exact title: 0
- Median DOL prevailing wage, level I / II: $68k / $89k

## Description
Are you someone who challenges the status quo, acts autonomously, and has experience leading platform and knowledge engineering work in broad-scale transformation programs? Are you ready to take the next leap in your career by leveraging your experience to convert a legacy software development process into one led by agentic models? Are you ready to build the tooling and knowledge infrastructure that other engineering teams will depend on every day? If the answer to all of the above questions is “Yes,” come join the world's leading professional services firm. If you are prepared and poised to take the next step in your career, you can help build the platform that powers agentic software delivery across a global business. Then we want to talk to you.
Work you’ll do
As a Manager, Agentic Engineering Platform, you will build and lead a small, high-leverage engineering team responsible for the infrastructure that underpins our agentic software development lifecycle. You will take a hands-on approach to designing and building the LLM Wiki and AWS Bedrock-managed knowledge bases that serve as steering content for agentic development tools. You will own the solution architecture behind these systems, applying agentic AI patterns such as retrieval-augmented generation, tool calling, and stateful workflows, and building in LLM evaluation, observability, and cloud-native deployment from the outset, all while keeping the platform aligned with enterprise architecture, security, compliance, and operational standards. You will build integrations between software development tools such as AWS Kiro and Azure DevOps, so that knowledge captured in ADO wikis becomes steering content, and so that agentic tools can read program artifacts such as epics, features, and user stories to generate code directly from approved requirements. You will also build integrations between ServiceNow and Azure DevOps that create defects automatically and route pull request assignments to the right developers. Beyond these integrations, you will deliver tools, utilities, and workflows that help technical program managers identify risk and cross-project dependencies by intelligently processing information across the platforms and tools our programs already use. The ideal candidate is a dependable team player and mentor who can move fluidly between hands-on engineering and defining the technical direction of a growing platform. Key responsibilities include the following:
Platform Ownership: Lead the architecture, build, and ongoing support of the LLM Wiki, AWS Bedrock-managed knowledge bases, and related steering content that power agentic development tools.
Agentic Solution Architecture: Design scalable, secure, and maintainable AI solutions using retrieval-augmented generation, tool calling, and stateful workflows, with LLM evaluation, observability, and cloud-native deployment built in.
Enterprise Alignment and Governance: Define service boundaries and integration patterns, and ensure the platform aligns with enterprise architecture, security, compliance, and operational standards.
Tool Integrations and Automation: Build and maintain integrations across AWS Kiro, Azure DevOps, and ServiceNow so program artifacts, defects, and pull request workflows move automatically through the platform.
Risk and Dependency Enablement: Deliver tools and workflows that help technical program managers identify risk, dependencies, and delivery issues across platforms.
A successful candidate would possess these skills:
Strong collaboration and relationship-building skills
Clear, concise communication with technical and non-technical audiences
Adaptability and a strong bias for learning and continuous improvement
The team
At Deloitte Tax LLP, our Product Engineering team within Global Employer Services (GES) Technology, has modernized software and product delivery, creating a scalable, cost-effective model that focuses on value and outcomes through a progressive and responsive talent structure. As Deloitte's primary internal development team, Product Engineering delivers innovative digital solutions to businesses, service lines, and internal operations with proven bottom-line results. Within Product Engineering, this team owns the agentic development platform: the LLM Wiki, AWS Bedrock-managed knowledge bases, and the integrations that connect our software development tools, program management systems, and ticketing platforms. Engineering teams across the organization depend on the steering files, knowledge bases, and integrations this team builds and maintains to run their agentic development workflows.
Qualifications
Required:
Ability to perform job responsibilities within a hybrid work model that requires US Tax professionals to co-locate in person 2 – 3 days per week
Bachelor’s degree in computer science, information technology, software engineering, or a related field.
5+ years of experience developing and deploying solutions using modern programming languages and cloud-native platforms (for example, Python, Node.js, TypeScript, SQL/NoSQL, AWS).
Strong solution design and architecture capabilities, with experience in Python agentic AI patterns, LLM evaluation patterns, retrieval-augmented generation, tool calling, stateful workflows, observability, and cloud-native deployment.
Ability to design scalable, secure, maintainable, and reusable AI solutions; define service boundaries and integration patterns; establish evaluation and reliability frameworks; and ensure alignment with enterprise architecture, security, compliance, and operational standards.
Hands-on experience building or integrating with large language model tooling, including retrieval-augmented generation, managed knowledge bases, and prompt or steering-content design (AWS Bedrock or equivalent).
Demonstrated experience building integrations between enterprise development, program management, and ticketing platforms (for example, Azure DevOps, ServiceNow, Jira, Confluence) using 

> History, not a promise: past H-1B filings do not mean an employer will sponsor you or a given role. Data: U.S. DOL OFLC LCA disclosure (FY2024–FY2026 for role counts, certified LCAs) and the USCIS H-1B Employer Data Hub. Not legal advice. Methodology: https://visajobs.io/methodology
