Elevating Aviation Regulatory Services and Public Engagement at KAA through an AWS-native Agentic and Generative AI Assistant
About the Customer
The Kenya Airports Authority (KAA) is a state corporation established to plan, develop, manage, regulate, and operate a safe, economically sustainable, and efficient civil aviation system in Kenya. Operating within the public sector and government enterprise segment, KAA is responsible for aviation safety and security oversight, provision of air navigation services, and the training of aviation personnel. As the ultimate regulator of Kenyan airspace, KAA serves a diverse stakeholder base, including commercial airlines, cargo operators, licensed pilots, drone operators, aviation engineers, and the traveling public.
Customer Challenge
KAA faced a critical operational challenge in managing public inquiries, licensing procedures, and regulatory guidance dissemination. Stakeholders frequently flooded manual support desks, email queues, and phone lines with high-frequency queries regarding drone registration (RPAS) guidelines, pilot licensing requirements, aviation medical exams, and real-time flight safety bulletins.
Because aviation regulations are dense, highly technical, and constantly updated, manual agents struggled to deliver instant, consistent, and legally accurate answers. This created severe operational bottlenecks, leading to inquiry response delays of several hours to days, increased compliance risks from outdated information, and heavy administrative strain on specialized safety officers. KAA required a highly available, secure, and context-aware conversational platform capable of serving thousands of concurrent international and domestic stakeholders 24/7, without compromising data security or regulatory precision.
Partner Solution
Coresys Solutions designed and deployed a secure, multi-channel Generative and Agentic AI Platform on AWS to automate regulatory guidance and simplify stakeholder interaction. The intelligence engine is powered by Amazon Bedrock (leveraging the Claude foundation model), which provides context-aware reasoning and natural language processing. To guarantee absolute factual accuracy, we built a Retrieval-Augmented Generation (RAG) architecture. This pipeline indexes dense regulatory publications, civil aviation acts, and licensing FAQs stored in Amazon S3, utilizing Amazon OpenSearch Service (with vector embeddings and k-nearest neighbor search) to ground every AI response in authorized KAA documents.
The solution is orchestrated using a LangGraph-style agent pattern hosted on Amazon EC2 instances within private subnets of an Amazon VPC, configured with Auto Scaling to seamlessly handle sudden traffic spikes. External user traffic is safeguarded by AWS Web Application Firewall (WAF) and managed by an Elastic Load Balancer (ELB). To maintain low latency, Amazon ElastiCache (Redis) caches high-frequency regulatory queries, while Amazon SQS decouples background event-driven processes, such as sending notification updates. Amazon Transcribe and Amazon Translate were integrated to support voice-to-text queries and multilingual assistance for international operators. Strict public-sector security compliance was enforced using AWS KMS for data encryption at rest, AWS IAM role-based privileges, and AWS CloudTrail for continuous auditing.
90%+
Improvement in Response Time enabling instant access to vital aviation guidelines.
65%
of incoming stakeholder queries successfully resolved autonomously.
$145K
Generated in AWS ARR with Amazon Bedrock representing 25% of consumption
Results and Benefits
The production launch of the KAA AI assistant yielded exceptional operational benefits:
- Over 90% Response Time Improvement: The average response time for complex regulatory and licensing queries dropped from a baseline of several hours/days to 2–3 seconds, enabling instant access to vital aviation guidelines.
- Efficient Automation of Inquiries: The AI assistant successfully resolved 55% to 65% of incoming stakeholder queries autonomously, significantly reducing the volume of general inquiries requiring human intervention.
- Enhanced Regulatory Compliance & Trust: By utilizing Bedrock Guardrails and RAG, KAA ensured that all generated answers were accurately grounded in official civil aviation publications, eliminating hallucinations and ensuring safe, compliant public engagement.
- Production-Grade Scale and Support: The system operates as a live, highly resilient workload driving $145,000 in AWS ARR (with Bedrock representing 25% of the total), fully demonstrating the viability of AWS-managed AI services in highly regulated government environments.
About the partner
Coresys Solutions Limited is a pioneering AWS consulting partner based in Nairobi, Kenya, specializing in cloud migration, application modernization, and advanced AI/ML solutions. We empower public sector organizations, educational institutions, and enterprises across East Africa to design, secure, and scale production-grade workloads on AWS. Coresys Solutions holds deep technical expertise in AWS-native security, serverless compute, and Generative AI architectures, serving as a trusted advisor for organizations seeking to achieve business-led digital transformation.