Scaling Student Support at Kenyatta University with a Multilingual AI Chatbot on AWS
About the Customer
Kenyatta University is one of Kenya’s largest and most prestigious public universities, headquartered in Nairobi with a student population running into the tens of thousands across its main campus, satellite campuses, and digital learning programs. Operating within the public higher education sector, the university offers a comprehensive range of undergraduate and postgraduate programs and attracts students from across Kenya, the East African region, and beyond. As a leading national institution with a large, diverse, and geographically distributed student community, Kenyatta University requires digital service channels capable of operating at a scale far beyond that of a typical institution.
Customer Challenge
The scale of Kenyatta University’s student population created a proportionally massive inquiry burden. Every admission cycle, national placement season, registration window, and examination period generated tens of thousands of repetitive questions — covering admission requirements, fee structures, hostel booking, unit registration, and academic calendars — directed at administrative offices through calls, emails, and physical visits. With a student community this large and diverse, staff could not respond consistently or quickly, and students in remote areas or those more comfortable in Kiswahili or other languages faced additional barriers to getting answers.
In the short term, the university risked long queues, frustrated students and parents during high-stakes windows such as first-year reporting, and administrative teams consumed by repetitive queries. In the long term, a manual support model that scaled only by adding staff was financially unsustainable for a
public institution, and inconsistent service quality risked eroding the student experience at one of the country’s flagship universities.
Partner Solution
Coresys Solutions designed and deployed an AI-powered FAQ Chatbot for Kenyatta University on AWS, architected to serve a national-scale student population. Amazon Bedrock powers the chatbot’s AI-driven responses, grounded in an institutional knowledge base stored in Amazon S3 to ensure answers reflect verified university information. To serve the university’s linguistically diverse community, Amazon Translate enables multilingual interactions and Amazon Transcribe supports speech-to-text queries, allowing students to engage with the assistant by voice and in the language they are most comfortable with.
The solution runs on a secure two-tier architecture within an Amazon VPC: an Elastic Load Balancer (ELB) in a public subnet, protected by AWS Web Application Firewall (AWS WAF), routes traffic to dedicated Amazon EC2 application servers in private subnets — one hosting the chatbot application and another hosting the administration portal — with a relational database isolated in its own private subnet. Amazon EC2 was deliberately selected over containers and serverless functions to provide the persistent, continuously running compute environment required for backend processing, administrative access, and tight orchestration across multiple integrated services. Amazon ElastiCache caches high-frequency queries for fast responses, while Amazon SQS manages asynchronous message processing to keep the platform responsive under peak load.
Deployments are fully automated through a modern CI/CD pipeline using GitHub Actions with OIDC federation into AWS IAM and AWS CodeDeploy, eliminating long-lived credentials and enabling controlled, repeatable releases. Amazon CloudWatch alarms and monitoring — including cost monitoring and alerting — provide proactive management of system health and ensure the environment remains performant and cost-efficient.
1–2 sec
Average response times for citizen inquiries
55-65%
Routine inquiries resolved autonomously
$95,000
in AWS ARR
Results and Benefits
The chatbot fundamentally changed how one of Kenya’s largest universities absorbs student demand:
- Peak-Season Resilience at National Scale: During high-pressure windows — national placement announcements, first-year reporting, and registration deadlines — the chatbot absorbed surges of concurrent inquiries that would previously have overwhelmed phone lines and front offices, keeping average response times at 1–2 seconds even at peak load and ensuring no student was left waiting during the moments that matter most.
- Inclusive Access for a Diverse Student Body: Multilingual and voice-enabled support through Amazon Translate and Amazon Transcribe removed language and accessibility barriers for a student community drawn from across Kenya and the region — extending equal, instant support to students regardless of location, language, or mode of interaction.
- Sustainable Support Economics for a Public Institution: By autonomously resolving 55% to 65% of routine inquiries, the university broke the link between student population growth and support headcount — a critical outcome for a public institution — while freeing administrative staff to handle complex, high-value student cases.
- Operational Agility and Cost Control: The automated GitHub Actions–CodeDeploy pipeline enables the university to ship knowledge base updates and new features in minutes rather than days, while CloudWatch cost monitoring and alerting keeps the workload cost-efficient. The solution operates as a live production deployment driving an estimated $95,000 in AWS ARR (with Amazon Bedrock representing 25% of consumption).
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.