• Own and maintain prompt engineering strategies, including prompt versioning, testing, and optimization • Design AI workflows that combine models, prompts, tools, enterprise data, and business logic • Implement AI orchestration layers to manage multi-step reasoning, decisioning, and actions • Ensure AI systems integrate cleanly into business workflows, APIs, and user interfaces • Apply guardrails to ensure safe, explainable, and compliant AI behavior • Support in building and maintaining production-grade deployment pipelines for AI solutions • Ensure reliability, scalability, cost control, and latency optimization • Implement monitoring and observability for AI systems (usage, performance, drift, failures) • Define and enforce change control, versioning, rollback, and release management processes • Collaborate closely with data scientists, actuaries and other business functions • Validate model behavior, outputs, and assumptions from a production and business-use perspective

• Design, develop, and maintain MCP servers that expose external tools and capabilities to AI assistants. • Implement specific MCP tools (functions) for reading files, managing issues, interacting with databases, or triggering complex workflows. • Define and structure resources, schemas, and prompts accessible to AI agents. • Ensure security and access control for the MCP server, limiting which tools and data agents can execute or access. • Manage API integrations and external data sources used by MCP tools. • Use JSON Schema or similar standards to document and validate tool definitions. • Integrate MCP Server tools with Azure AI Foundry agents, enabling AI agents to perform actions such as data querying, workflow automation, and API interaction. • Collaborate with product managers and front-end teams to integrate AI capabilities into user-facing platforms (e.g., Microsoft Copilot, VS Code extensions, or web portals). • Utilize Azure Functions, Azure Logic Apps, and Azure Cognitive Services to connect MCP-based AI workflows to enterprise systems. • Develop and manage tools that interface with Azure resources (e.g., Azure Storage, Azure SQL, Azure Compute Services) through the MCP Server. • Enable AI agents to securely manage, retrieve, and utilize Azure resources. • Apply best practices in cloud security, authentication, and API permissions management. • Conduct comprehensive testing and debugging of MCP tools and Azure integrations. • Create detailed documentation for MCP tools, including usage examples, API specifications, and integration guides. • Ensure scalability, performance, and reliability of MCP and AI Foundry deployments. • Partner with AI engineers, cloud architects, and data scientists to align tool development with strategic goals. • Present technical solutions and architectural decisions to stakeholders in clear, actionable terms. • Contribute to the AI governance framework, ensuring responsible AI tool design and usage.

• Design, build, and ship production-grade GenAI features that help teams explore data, ask better questions, and trust the answers they get. • Own GenAI features end-to-end — from system design and LLM integration through deployment, monitoring, and iteration. • Build and maintain reusable GenAI platform components that make it easier and safer for teams across Teamworks to ship AI-powered features. • Establish and document best practices for building, testing, and operating GenAI systems reliably in production. • Enable product engineering teams to ship AI-powered features by providing shared infrastructure, technical guidance, and hands-on integration support. • Make architectural decisions that balance speed, scalability, reliability, security, and cost in a multi-tenant data environment. • Partner with Product, Design, and domain subject-matter experts to deliver trustworthy, explainable AI experiences that align with real user workflows.

• Design multi-agent architectures with robust state management, memory, and routing. • Choose and implement leading frameworks such as LangGraph/LangChain Agents, Microsoft AutoGen, CrewAI, LlamaIndex Agents, Semantic Kernel, or Haystack Agents—and justify trade-offs. • Build modular components (planners, tool registries, policy guards, evaluators) that are reusable across clients and domains. • Integrate enterprise tools and data sources via function/tool calling, webhooks, and event-driven flows (Queues/Service Bus/Functions). • Implement retrieval-augmented generation (RAG) patterns with vector stores (Azure AI Search, pgvector, MongoDB Atlas, Pinecone, Weaviate, Milvus) and structured knowledge (SQL/Graph). • Add deterministic fallbacks, circuit breakers, and caching to keep latency and cost predictable. • Define SLIs/SLOs for agent runs; implement tracing, metrics, and logging (e.g., Langfuse + OpenTelemetry) and build dashboards for run-level analytics. • Create evaluation harnesses (automatic + human-in-the-loop) using tools such as Ragas, DeepEval, promptfoo to measure groundedness, task success, safety, and cost. • Productionize with CI/CD, environment promotion, feature flags, and canary strategies; optimize cost-per-task and time-to-success. • Enforce content and safety policies (redaction, classification, guardrails) with policy-as-code; implement role/tenant isolation and data minimization. • Collaborate with security teams to align to ISO 27001/SOC 2/NIST/HIPAA/GDPR contexts; deliver audit-ready evidence for agentic workflows. • Build privacy-first patterns (no data exfiltration by default, least-privilege tool access, secure prompt/trace storage). • Work directly with enterprise client teams to translate business processes into agentic designs; present trade-offs and proofs-of-value that lead to production. • Partner with solution leads to create domain-specific agents (e.g., RFP assist, incident RCA drafting, knowledge ops) and reusable templates.