Director – AI & Data Architecture (Pre-Sales & Solutioning)

What matters most

• Shape enterprise pursuits

  The role partners with sales and account teams to qualify opportunities, leads executive discovery and AI strategy discussions, and turns business challenges into solution roadmaps for strategic and large, multi-million-dollar pursuits.

• Architect modern data platforms

  CGI calls for enterprise designs spanning AWS, Azure or Google Cloud, Snowflake, Databricks, lakes or lakehouses, warehousing, ETL/ELT, streaming, governance and master data management.

• Make AI designs usable at enterprise scale

  The director designs solutions involving generative and agentic AI, RAG, LLMs, assistants, machine learning and automation, while evaluating emerging technology and developing reusable accelerators and reference architectures.

• Connect executive advice to delivery

  The role presents recommendations to business and technology leaders, supports proposals and RFP responses, and works with delivery teams to transition solutions into implementation.

• Raise architecture practice standards

  CGI expects the director to mentor architects and engineers and promote architecture standards, governance and engineering best practices.

Likely challenges to think through

• Commit to a solution before every detail is known

  Likely preparation challenge, not a stated CGI problem: pre-sales pursuits require a credible roadmap while discovery may still be incomplete. Prepare an example of how to surface assumptions, qualify an opportunity and distinguish a proposed architecture from decisions needing validation.

• Balance AI ambition with enterprise controls

  Likely preparation challenge: selecting among RAG, agentic AI and other approaches while accounting for the posting’s security, governance and enterprise-architecture requirements. Prepare to explain when a simpler approach is preferable and how its data dependencies would affect the design.

• Balance tailored proposals with implementable standards

  Likely preparation challenge: a client-specific proposal must still be feasible for delivery teams and consistent with reusable reference architectures. Prepare an example showing how you documented trade-offs, preserved room for client needs and made a handoff actionable.

Topics to prepare

• Enterprise AI and data opportunity shaping

  Review how to qualify a business challenge, define outcomes and scope, and turn discovery findings into an implementation roadmap. Practice outlining the decisions needed for a hypothetical executive discovery session. Relevance: the role leads workshops and shapes strategic, large-scale pursuits with sales and account teams.

• Cloud data platform architecture

  Compare when a data warehouse, lake or lakehouse is appropriate, including the roles of Snowflake, Databricks and AWS, Azure or Google Cloud. Practice explaining the trade-offs for one hypothetical enterprise data modernization scenario. Relevance: the posting explicitly calls for cloud platform design and hands-on Snowflake experience.

• Data integration and analytics patterns

  Revisit ETL versus ELT, streaming versus batch ingestion, and how integration choices affect scalable analytics. Practice sketching two ingestion options and explaining the trade-offs. Relevance: designing scalable ETL/ELT, analytics and data integration solutions is required.

• Generative AI solution architecture

  Compare LLM, RAG, agentic AI and AI-assistant architectures, including where each fits an enterprise use case. Practice explaining a proposed architecture and its limitations to a business stakeholder. Relevance: the role designs these solutions and evaluates emerging AI technologies.

• Governance, security and architecture standards

  Review how data governance, master data management and security influence an enterprise architecture. Practice identifying governance decisions and delivery-team handoff questions for a hypothetical design. Relevance: these principles are required knowledge, and the director promotes architecture standards.

Skills to revise

• Facilitate executive discovery

  As a preparation exercise, draft a short workshop agenda and questions that move from business objectives to data constraints, AI opportunities and next decisions. Relevance: the role facilitates executive workshops, discovery sessions and AI strategy discussions.

• Develop a Snowflake-centered solution design

  Practice drawing a cloud data platform design that shows ingestion, transformation, storage and analytics around Snowflake; explain where another stated technology, such as Databricks, would or would not help. Relevance: hands-on Snowflake experience and modern cloud data platform architecture are explicit.

• Produce a solution roadmap and delivery handoff

  Practice turning a hypothetical architecture into phased implementation steps, dependencies, decisions and a concise handoff for engineers. Relevance: the posting calls for implementation roadmaps and partnership with delivery teams.

• Present and defend a proposal

  Practice a brief executive presentation of a proposed AI and data solution, then answer questions about trade-offs, governance and implementation. Relevance: the director advises stakeholders and supports proposals, RFP responses and executive presentations.

• Map the stated requirements to evidence

  Prepare a one-page inventory of relevant architecture experience, including dates for the stated 12+ and 5+ year requirements, degree, pre-sales work, cloud expertise, hands-on Snowflake work, and ETL/ELT and integration designs. Note which examples you can discuss in detail.

• Build a pursuit walkthrough

  Select a real, shareable pre-sales or consulting example. Outline how you qualified the opportunity, ran discovery, shaped the solution, contributed to a proposal or RFP response, and explained trade-offs to executives; use actual outcomes only.

• Practice an executive discovery session

  Draft a short workshop agenda and questions that uncover business goals, data constraints, security needs, stakeholders, and success measures. Rehearse turning the answers into an initial solution scope and roadmap; this is preparation for the role’s stated workshop duties, not a stated interview format.

• Sketch a cloud data-platform design

  Prepare a diagram and brief walkthrough using a cloud provider you know. Show where Snowflake, ingestion and ETL/ELT, analytics, and—where appropriate—streaming or lakehouse components fit, with governance, security, scale, and integration decisions explained.

• Work through an enterprise AI use case

  Choose a plausible client problem and sketch when RAG, an LLM assistant, agentic AI, or conventional ML would be appropriate. Explain data access, evaluation, security controls, and the boundary between a reusable reference architecture and client-specific design.

• Prepare a delivery and leadership handoff

  Create a sample roadmap with phases, dependencies, risks, and decisions for a delivery team. Add a brief example of how you have mentored architects or promoted standards, if you have one you can substantiate.

• Write targeted questions for CGI

  Bring questions about how sales, architecture, and delivery teams divide responsibilities; what makes an AI or data pursuit worth qualifying; how reusable accelerators are governed; and what a successful solution handoff looks like.