Technical Lead / Architect

A senior architect who sets the technical direction for the capability. You will design the reference models for each type of AI build, set engineering standards, and serve as engineering lead for a small team of contractors.

What you will do

· Design reference architectures for each type of AI build, from short-lived personal tools to long-lived, client-facing solutions and autonomous agents.

· Define engineering standards, templates, and controls so builds are secure and supportable.

· Lead the engineering team day to day by planning the work, reviewing designs and code, and clearing blockers.

· Design the AI agent’s recommendation logic, guardrails, and rules for when to start setup or route to a manual intake.

· Map needs to existing firm platforms first, and recommend new components only where they are justified.

· Work with firm platform, security, and risk teams to get designs reviewed and approved.

· Support each wave’s pilot and go-live, and use builder feedback to shape the next wave.

· Document architecture decisions and hand over to the support team at the end of the engagement.

· 14+ years in software engineering or solution architecture, including 3+ years designing solutions on Microsoft Azure.

· Hands-on experience architecting generative AI solutions in production, such as retrieval, tool use, and agents, with models from Anthropic (Claude), OpenAI, or Google (Gemini).

· Experience with agent frameworks such as the Claude Agent SDK, Google Agent Development Kit (ADK), or Microsoft Foundry.

· Experience with cloud security, identity, and data protection for confidential or regulated data.

· Experience defining reference architectures, engineering standards, or internal developer platforms.

· Experience leading small engineering teams, including design and code reviews.

· Proficiency in Python or TypeScript, and working knowledge of infrastructure as code and GitHub Actions.

· Able to explain technical choices to senior, non-technical stakeholders.

Depends on Experience

Cyber & Tech Risk (CTR) practice is building an AI-native build capability made up of reference models (ready-made kits) for common types of AI solutions, an AI agent that points builders to the right kit and starts simple setup, and shared components in a central code repository. The aim is to get AI solutions into production faster and with less rework. The program delivers in three waves, going live in January, April, and September 2027, on Microsoft Azure and AI platforms from Microsoft, Anthropic, OpenAI, and Google.