Meet Our Customer

Our customer is one of Saudi Arabia's largest and longest-established grocery retail chains, operating a nationwide network of stores and serving millions of customers across the Kingdom.

As the business grew, so did its data. Every day its stores generate an enormous volume of operational detail — sales, inventory, stock movement, promotions and procurement — flowing from SAP S/4HANA into an Amazon Redshift data warehouse that now holds more than 600 million transaction line items and 125 million baskets, growing by over half a million line items each day. The company had built a genuine enterprise data asset. What it wanted next was to put that asset directly into the hands of the people running the business, in the language they think in, at the moment they need it.

The Challenge

The data existed. Access to it did not.

Store managers, category buyers and executives depended on BI dashboards and analyst queues for even routine questions — how a store performed yesterday, which products were running low, whether a promotion moved volume. Answers took hours or days, arrived in English only, and required desk access. A store manager and a member of the executive team waited in the same report queue.

At the same time, the highest-volume operational processes remained manual. Daily purchase-order creation for fresh and non-fresh goods, demand forecasting, competitor price tracking and customer win-back campaigns were spreadsheet-driven and dependent on a small group of specialists — work that consumed skilled hours without scaling.

The customer set NorthBay a deliberately demanding brief: a single, trustworthy assistant that employees could simply talk to. It had to understand retail language in both Arabic and English, answer from governed enterprise data rather than guesswork, respect each employee's role and permissions, and safely take action — not just report on it. It had to serve every persona in the business, from the store floor to the boardroom, run inside the tools employees already use, and scale from a first release to the entire organisation.

The Vision

NorthBay designed an agentic AI platform running on Amazon Bedrock inside the customer's own AWS environment, so enterprise data stays under the customer's governance and every persona receives a different, role-appropriate answer from the same trusted foundation.

A supervisor-and-specialist agent architecture

Rather than a single monolithic bot, NorthBay built a multi-agent system on Amazon ECS Fargate in which a supervisor agent interprets each request and routes it to one of nineteen specialists — sales analytics, inventory, stock movement, sales budgets, product catalogue, competitor pricing, predictive forecasting, purchase-order creation and inquiry, HR knowledge, IT incidents, business events, weather and scheduled reporting among them. Each specialist pairs model reasoning with governed tools: text-to-SQL over the Redshift warehouse, retrieval-augmented generation over HR and policy documents on Amazon Aurora Serverless v2 with pgvector, ServiceNow ticketing, SAP purchase-order APIs, and Airflow job scheduling on Amazon MWAA. Agents either complete their task or escalate back to the supervisor, so a conversation can move across domains without the employee needing to know which system holds the answer.

Reasoning that holds up against a real enterprise schema

The hard part of retail analytics is not generating SQL — it is generating correct SQL against a warehouse that encodes years of business rules. Anthropic's Claude models on Amazon Bedrock translate ambiguous, bilingual retail questions into precise queries across 130+ tables, respecting posting versus transaction dates, fresh versus non-fresh classification and store hierarchies. Vision capability reads product photos, shelf images and competitor brochures; voice queries flow through Amazon Transcribe into the same agents. One assistant detects and responds in each user's language, so Arabic and English speakers share a single platform rather than two.

Governed by design

Amazon Bedrock Guardrails, role-based access control enforced at the agent layer and prompt-injection hardening keep responses grounded and workplace-appropriate. The platform is deployed with infrastructure-as-code, CI/CD with environment promotion and explicit production-deploy authorisation, Amazon SageMaker AI pipelines for demand forecasting with SHAP explainability, and event-driven integration to SAP and messaging channels. Security follows an enterprise programme of IAM least-privilege, KMS encryption, managed secrets and OIDC-only pipelines, with PDPL-aligned data handling and full auditability of agent actions.

Outcome

The platform has been in production since 2025, delivered through four major releases and two funded extension phases, with a third in planning — sustained commitment rather than a one-off pilot. It runs across development, QA and production environments with formal release management, a 12-week software-quality-assurance cycle and hotfix-governed production freeze discipline. It is delivered inside Microsoft Teams, with an embedded web chat option, so adoption required no new application: employees message the assistant like a colleague.

The change is as much economic as technical:

• Self-service analytics. Questions that previously entered a reporting queue are answered conversationally, removing a dependency that constrained decisions at the point of operation.
• Procurement automation. Purchase orders are created automatically every morning at 08:30 KSA, integrated end to end with SAP, replacing a manual spreadsheet process and returning buyer hours to negotiation and assortment strategy.
• Forecasting with accountability. SageMaker demand forecasts with SHAP explainability inform ordering, with drivers — events such as Ramadan and Eid, weather, promotions — visible to the business rather than buried in a model.
• Enterprise data at conversational reach. Records that were the preserve of data specialists are now open to any authorised employee, under the same governance.
• Measurable internal communication. Organisation-wide announcements carry per-recipient delivery tracking, on a pipeline validated at 1,000+ recipients.

Value across the business

Because the supervisor-agent architecture separates routing from expertise, the same platform serves each function differently — without separate tools or integrations to manage. Six of seven functions are live in production today.

AWS Services

The project heavily utilized AWS services to ensure scalability and efficiency:

Why NorthBay?

From the outset this was built for daily operational use across an entire workforce — which meant release management, security review, bilingual quality assurance, and the discipline to ship one use case, prove it in production, and expand from there. NorthBay is an AWS Premier Consulting Partner with an AI Consulting Services Competency and deep delivery experience in the region, and applied the same engineering rigour to an agentic system that it would to any enterprise platform: governed data access, explainable automation, and a production operating model from the first release.

Procurement data management, often seen as a necessary but overlooked function within organizations, can be transformed into a strategic asset that drives significant value. With the increasing availability and affordability of data, procurement can now be used to improve cost savings, mitigate risks, enhance decision-making, improve compliance, and gain a competitive advantage.

The Value of Procurement: By effectively analyzing and interpreting procurement data, organizations can gain valuable insights into their procurement processes and identify areas for improvement.

Cost Savings: Procurement data enables opportunities for price negotiation, bulk purchasing, and consolidation. It also highlights inefficiencies, such as duplicate payments or unauthorized spending, helping reduce waste.

Risk Mitigation: By analyzing procurement data, businesses can monitor supplier performance, identify potential risks, and anticipate supply chain disruptions before they occur.

Informed Decision-Making: Leveraging procurement data allows for better strategic sourcing and supplier selection. It also supports more accurate decisions around contract terms, pricing structures, and payment options.

Enhanced Compliance: Procurement data ensures regulatory and policy compliance by tracking supplier certifications and approvals, safeguarding against non-compliant transactions.

Strategic Advantage: By leveraging procurement data effectively, organizations can gain a competitive edge through optimized procurement processes and identify opportunities for innovation and cost reduction.

To realize the full potential of procurement data, organizations must implement effective data management practices, including:

• Data Quality: Ensuring data accuracy, completeness, and consistency.
• Data Governance: Establishing clear policies and procedures for data collection, storage, and access.
• Data Analytics: Utilizing advanced tools and techniques to extract meaningful insights from data.
• Integration: Integrating procurement data with other systems (e.g., ERP, CRM) for a holistic view of the business.

Data Cleansing and Standardization: Ensuring data quality through cleaning and standardization processes.

• Data Visualization: Using visual tools to communicate complex data insights effectively.
• Data-Driven KPIs: Developing key performance indicators (KPIs) to track procurement performance and identify areas for improvement.
• Continuous Improvement: Implementing a culture of continuous improvement and data-driven decision-making.

Several emerging technologies and trends will shape the future of procurement data, including:

• Artificial Intelligence (AI): Leveraging AI for tasks such as contract analysis, supplier risk assessment, and predictive analytics.
• Blockchain: Utilizing blockchain technology to improve supply chain transparency and traceability.
• Internet of Things (IoT): Collecting and analyzing data from IoT sensors to optimize procurement processes and inventory management.
• Advanced Analytics: Employing advanced analytics techniques (e.g., machine learning, deep learning) to uncover hidden patterns and insights in procurement data.

Procurement data processing and analytics can benefit organizations in a wide range of industries, including:

• Manufacturing: Optimizing inventory levels, managing supplier relationships, and identifying cost-saving opportunities.
• Healthcare: Improving supply chain efficiency, reducing drug costs, and ensuring compliance with regulations.
• Retail: Enhancing demand forecasting, optimizing product assortment, and negotiating better terms with suppliers.
• Technology: Managing complex supply chains, ensuring component availability, and controlling costs in a rapidly evolving market.
• Government: Improving public procurement processes, reducing waste, and ensuring transparency and accountability.

In today’s rapidly evolving business landscape, manufacturers face a constant challenge: maintaining a resilient supply chain. Disruptions can arise from various sources, including natural disasters, unexpected market shifts, and global events. These disruptions can lead to delays, stockouts, and ultimately, dissatisfied customers.

However, there’s a powerful Generative AI-based solution to help you navigate these challenges: Amazon Q Business. This powerful AI assistant can empower manufacturers to enhance their supply chain management and mitigate risks.

Amazon Q Business is a state-of-the-art generative AI assistant that harnesses the capabilities of advanced language models and machine learning algorithms. It provides a unified search experience across enterprise data, helps users with analyzing it, generates permissions-aware responses, and provides accurate recommendations in real-time. Its ability to understand context, learn from interactions, and adapt to specific business needs makes it an invaluable tool for enhancing supply chain resilience.

Key Benefits of Amazon Q Business:

• Enhanced Supply Chain Visibility

  : Gain real-time insights into your entire supply chain operations, allowing you to identify potential disruptions proactively.

• Risk Mitigation: Utilize AI-powered analytics to assess supplier risks, forecast demand, and plan for contingencies.
• Streamlined Operations: Automate routine tasks like order tracking, inventory management, and vendor communication, freeing up valuable time for strategic decision-making.
• Data Security: Rest assured that your sensitive data is protected with Amazon Q Business’s robust security measures.
• Customization: Tailor the platform to your specific needs by integrating it with your existing enterprise systems.

How Amazon Q Business Works:

• Data Integration: Connect your supply chain data to Amazon Q Business, including information from S3, ERP systems, CRM platforms, and other relevant sources.
• Query and Analysis: Ask questions in natural language via chat interface and receive AI-generated responses based on organization data.
• Actionable Insights: Gain valuable insights to optimize your supply chain, reduce costs, and improve customer satisfaction.

One of the key advantages of Amazon Q Business is its ability to support improved visibility into supply chain operations. By leveraging the data and information available within an organization’s systems, Amazon Q Business can assist manufacturers in gaining a more comprehensive view of their supply chain ecosystem when queried.

Use Cases

• Assisting with Potential Disruption Analysis:

  When provided with relevant data, Amazon Q Business can help analyze information about global events, weather patterns, and supplier performance to assist manufacturers in identifying potential disruptions. For instance, it could help interpret reports on global events that might affect shipping routes or analyze supplier performance data to highlight potential issues.

• Summarizing Supply Chain Reports: Instead of spending hours poring over lengthy reports, managers can ask Amazon Q Business to summarize key findings and highlight critical issues from the documents they provide. This allows for more efficient information processing and can support quicker decision-making.
• Providing Insights Based on Available Data: Amazon Q Business can help analyze provided historical data and current information to offer strategic insights to optimize supply chain performance. This might include helping to interpret data for inventory reallocation, alternative sourcing options, or process improvements.

Amazon Q Business can play a supportive role in improving the accuracy of demand forecasting and planning. When given access to historical sales data, market trends, and information on external factors such as economic indicators or seasonal variations, it can assist in generating more informed demand forecasts. This enables manufacturers to make better-informed decisions about optimizing inventory levels, managing carrying costs, and improving product availability.

Additionally, when provided with relevant data, Amazon Q Business can assist in supplier risk assessment by helping to interpret information on factors such as financial health, delivery performance, and quality metrics. This approach allows manufacturers to gain insights that can help identify potential issues and support decision-making on measures such as diversifying suppliers or adjusting order quantities to maintain supply chain stability.

Data has become an invaluable asset in today’s rapidly changing business world. Effective data analysis is no longer just a nice-to-have; it’s essential for making informed decisions and staying competitive. Small and medium-sized businesses (SMBs) are increasingly recognizing the power of data-driven insights. However, traditional business intelligence (BI) tools can be complex and time-consuming to use, making it difficult for SMBs to leverage their data effectively.

Amazon QuickSight Q is a GenAI-powered BI tool designed to simplify data analysis through natural language interaction. It allows users to ask questions about their data in plain English, eliminating the need for complex query languages or extensive technical knowledge.

• Key features of QuickSight Q include:
• Natural language processing for data queries
• AI assisted dashboard authoring
• Broad data source support
• Serverless auto scaling

By enabling business users to interact with data using everyday language, QuickSight Q breaks down the barriers between decision-makers and their data, fostering a culture of data-driven decision-making across the organization.

In an increasingly competitive marketplace, SMBs must make informed decisions quickly to stay ahead. Data-driven decision-making is no longer a luxury but a necessity for businesses for several reasons:

• Increasing competition: With the rise of digital transformation, SMBs face competition not just locally, but globally. To stay competitive, they need to make informed decisions quickly.
• Market volatility: Economic conditions and market trends can change rapidly. SMBs need to be able to analyze and react to these changes in real-time.
• Customer expectations: Modern consumers expect personalized experiences and quick responses. SMBs need data insights to meet these expectations effectively.
• Resource optimization: Unlike large enterprises, SMBs often have limited resources. Data-driven insights help allocate these resources more efficiently.
• Regulatory compliance: Many industries face increasing regulatory requirements. Data-driven approaches help ensure compliance and reduce risks.

While the need for data-driven decision making is clear, traditional BI tools present their own set of challenges that Quicksight Q was built to address:

• Seamless integration with AWS identity and access management
• Fine-grained access controls to manage user permissions
• Encryption of data at rest and in transit to protect sensitive information
• Comprehensive audit logs for monitoring and compliance purposes

In recent years, the Middle East and Africa (MEA) region has witnessed a surge in cloud computing adoption, particularly within the financial services sector. This technological shift promises numerous advantages for financial institutions, including cost savings, improved scalability, enhanced security, and faster innovation. However, navigating the complexities of cloud migration in MEA region requires careful planning and consideration of unique factors. How can these institutions maximize the benefits while minimizing risks? This post explores key strategies and best practices for cloud migration in the MEA financial services market.

The MEA region is a tapestry of nations, each with its own cultural, economic, and regulatory fabric. From the tech-savvy United Arab Emirates to the emerging markets of Nigeria and Kenya, the region spans a wide spectrum of technological development and market maturity. Other key players in this diverse landscape include Kuwait, Qatar, Bahrain, Saudi Arabia, Egypt, South Africa, Morocco, and Algeria.

Historically, cloud adoption in MEA was slow due to factors such as expensive and poor-quality international connectivity in some countries and regulatory uncertainty around data storage. However, recent government ambitions to diversify and digitize their economies have accelerated the introduction of cybersecurity, data protection, and cloud-specific policies and regulations.

This shift has resulted in an increasing number of international cloud service providers (CSPs) establishing a presence in the region. For example, AWS launched its first Middle Eastern cloud center in Bahrain in 2019 and expanded into the UAE in 2022. Microsoft entered the UAE in 2019 and established data centers in Qatar in 2022. Several more CSPs are expected to be operating in Saudi Arabia by 2030.

The largest source of cloud’s value-creation potential lies in its ability to accelerate product development and scale products efficiently. Cloud migration offers several key benefits for financial services in MEA:

• Improved Security and Adherence to Regulations: Cloud providers prioritize robust security measures to safeguard their infrastructure and data, ensuring compliance with industry standards and regulations. Financial institutions can benefit from these robust security standards, which often exceed what individual organizations can implement on their own, building trust with regulators, shareholders, and customers.
• Cost Efficiency: Cloud computing significantly reduces IT infrastructure costs by eliminating the need for on-premises hardware, software, and maintenance. This shift from capital expenditure to operational expenditure frees up resources for strategic investments and innovation.
• Improved Decision-Making: Cloud-based analytics and AI capabilities enable financial institutions to process vast amounts of data quickly and derive actionable insights. This can lead to more informed, data-driven decisions across all levels of the organization, from risk assessment to customer service strategies.
• Increased productivity: Cloud solutions offer collaborative tools and centralized data storage, enabling teams to work more efficiently. Automation of routine tasks and improved access to information can significantly boost employee productivity across departments.
• Faster time to market: The agility provided by cloud infrastructure allows financial institutions to develop and deploy new products and services more rapidly.
• Improved Scalability and Resilience: Cloud-based solutions can easily scale based on demand and provide robust disaster recovery options across multiple data centers, enhancing operational resilience.
• Agility and Innovation: The cloud enables faster innovation by providing access to cutting-edge technologies like AI, machine learning, and blockchain. This can help financial institutions stay ahead of the curve in developing new products and services, such as AI-powered robo-advisors or blockchain-based payment solutions.
• Enhanced Customer Experience: Leveraging cloud technologies, institutions can deliver personalized services, faster response times, and improved digital interactions, elevating customer satisfaction.

The public cloud offers significant value through advanced services that would be cost-prohibitive to develop independently. These include Software as a Service (SaaS) solutions, advanced analytics, and machine learning capabilities. The cloud also provides unlimited, on-demand, and scalable infrastructure capacity, storage, and computational power.

Financial institutions can leverage the cloud to generate value by:

These technologies, while still emerging, have the potential to revolutionize financial services in the coming years. The cloud provides a flexible, cost-effective platform for experimenting with and gradually implementing these innovations.

Imagine building a house on a shaky foundation. No matter how expertly you construct the walls or how beautifully you decorate the interior, the entire structure remains at risk. In the world of Generative AI, data quality is that foundation. As organizations rush to harness the power of Generative AI, many overlook this critical element, potentially jeopardizing their entire AI strategy. In this post, we’ll explore why data quality is vital to successful Generative AI implementation, what organizational changes are necessary to support it, and how AWS technologies can help you build a rock-solid data foundation for your AI initiatives. By the end, you’ll understand not just the ‘what’ of data quality for Generative AI, but the ‘why’ and ‘how’ that can set your projects apart.

Generative AI, much like traditional machine learning, adheres to the ‘garbage in, garbage out’ principle. This means the quality of AI-generated content is fundamentally tied to the accuracy of its training data. Similar to earlier machine learning systems, Large Language Models (LLMs) learn patterns and relationships within their training data. These learned patterns then form the basis for their output generation. However, this process is highly sensitive to data quality issues:

• Data Noise:

  At its most basic level, poor-quality data introduces noise into the trained model. This noise interferes with the model’s ability to discern meaningful patterns, leading to less accurate and less coherent outputs.

• Biased Outputs: More insidiously, flawed data can introduce bias into the model. This bias can manifest in various ways, from subtle shifts in language tone to more overt imbalances in representation or perspective. Bias data during training limit the model’s ability to generalize across diverse contexts and may lead to unfair or skewed outputs.
• Misinformation: In severe cases, training on inaccurate or false data can result in the model generating outright misinformation, potentially spreading falsehoods at scale.

The consequences of these data quality issues extend beyond mere technical inaccuracies. As users interact with AI systems built on flawed data, they gradually lose trust in the system’s reliability. This erosion of trust can have far-reaching implications:

• Diminished Utility:

  Users may hesitate to rely on the system for critical tasks, significantly limiting its practical usefulness.

• Reduced Adoption: A reputation for unreliability can hinder wider adoption of the AI system within an organization or among potential customers.
• Financial Losses: The costs associated with developing and deploying an AI solution that fails to meet user needs due to data quality issues can be substantial, representing a significant waste of resources.
• Reputational Damage: For businesses, consistently inaccurate or biased AI outputs can lead to reputational damage, potentially affecting customer relationships and market position.

Ensuring high-quality training data is not just a technical necessity, but a critical business imperative for any organization looking to leverage the power of generative AI effectively and responsibly.