12 Weeks to AI

Getting started on an Artificial Intelligence (AI) journey, implementing AI responsibly, and ensuring long-term meaningful adoption is challenging for all organizations. Successful AI implementations with AWS require educating stakeholders about AI’s potential and required guardrails, developing a cohesive AI strategy with the best practices, assessing organizational data maturity, and establishing a clear path to production with AWS and organization wide adoption at scale.

With over a decade of experience in AI (Artificial Intelligence), and multiple years of being a strong AWS partner, we combine our end-to-end capabilities in AI with deep domain knowledge and technology engineering to demystify and implement AI.

Our 12 weeks to AI approach is designed to be adaptable and flexible to meet your unique data maturity and business goals.

In these 12 weeks, utilizing AWS infrastructure and AI services, CGI will be a strong partner by your side to help you:

Strategically think about AI – Collaborate with your organization, scan industry landscapes, and bring in CGI best practices in AI built with AWS Machine Learning to co-generate use cases that could bring the most value for your business.

Prioritize use-cases that make most sense – Evaluate ML use cases against a set of defined criteria such as: technical feasibility, business value generated, fit with selected AWS AI services, fulfillment of your business goals, and competitive advantages achieved amongst other criteria.

Build low-cost AI models to test value of your data – Rapidly prototype a Proof of Value using a low-cost AI model, hosted on AWS AI Infrastructure, trained on a subset of your company data, which helps ascertain the technical feasibility and estimate business value of your top AI use-cases.

Create a roadmap to scale and maximize your AI investment – Based on the Proof of Value, the roadmap contains a clear path to scaling, production, includes best practices on data and AI, and helps you procure internal buy-in for a further investment into AI.

Clear benefits of this engagement:

Start Right from Where You Are

• Proven Data and AI assessments and AWS best practices to ensure you start right
• Top use cases mapped against tested criteria that focuses on value and feasibility

Build Responsibly and Re-use

• Responsible AI baked in, not bolted on
• Build reusable models focused on minimizing time to market and enabling scale

Test Before You Invest

• Build AI models to test the top use cases leveraging your actual data
• CGI’s iterative approach to AI model validation minimizes risk and cost

Expedite Buy-in

• Confidently provide answers to leadership of expected ROI
• Build your business case based on validated results and a clear vision to production

Detailed Phases of the 12 Weeks: In a typical engagement, a CGI squad of AI experts collaborates with your key stakeholders to co-generate Machine Learning solutions using AWS, following our AI best practices for responsible and ethical development.

Envision:

• Conducting strategic alignment workshops to explore and understand your current state and initial ambitions with AI
• Create a list of ML use-cases by adding from each of the steps below:
• Scanning the competitive landscape
• Looking for common use cases across product lines and specific use cases for each product line
• Bringing in logical and impactful use-cases suggestions from our CGI experts
• Conduct initial Responsible Use of AI assessment and Risk/Compliance assessment
• Utilizing AWS Cognitive components and accelerators

Explore:

• Evaluating and scoring each ML use case against a detailed framework of evaluation parameters
• Holding collaborative and iterative evaluation sessions with your Technology Leads, Product or Business Owners and Business Leadership and shortlist top use-cases to target

Build:

• Working with your technical teams to gather access to a subset of company data
• Quickly preparing a basic, typically optimal AWS AI infrastructure and environment to run the AWS Machine Learning model while ensuring security and privacy constraints
• Identify any correlations in data, gaps, or dependencies we need to watch for when scaling this use case

Mobilize:

• Creating a roadmap to scale the top ML use cases by leveraging CGI’s expertise and best practices
• Roadmap typically includes recommendations for best resources to ensure optimal performance and output, such as AWS File Storage Service or Databricks for data lakes, AWS Glue for defining and scheduling data pipelines, and plan for deploying use case at scale with Amazon Elastic Kubernetes Service (EKS)
• Getting alignment on the roadmap by coordinating with leadership team, business owners and technical owners