Key highlights
~ 50
faster student query response
30
quicker registration process
30
boost in student engagement
Challenges
Manual data preparation for RASA chatbots required significant labor and time.
Traditional support methods failed to address student queries about administration and courses.
Students struggled to navigate complex administrative processes and course details.
Solution
AWS-native zero-ETL pipelines, a three-tier Redshift architecture with data sharing, Airflow-based orchestration, automated MLOps runbooks, and cost-optimized DevOps
Zensar partnered with the customer team to operationalize an AWS-first architecture with Zero-ETL ingestion and curated Redshift layers. Data sharing bridged BI, EDL, and food demand clusters, giving analysts secure, near-real-time access without physical data movement. Airflow orchestrated multi-account pipelines with parameterization, alerts, and metadata-driven execution. MLOps standardized model deployment via a Kedro runbook, integrating CI/CD and operations. Return-leg pipelines captured forecasts from partners. EMR, Glue, and SQL jobs were optimized to cut execution times by 50%+. Cluster RPUs were right-sized (128 → 64), and lower environments were shut down when idle, achieving ~50% cost savings.
1.
The solution used a conversational interface that employed natural language processing (NLP) to understand student queries and provide relevant answers.
2
The gen AI model retrieved accurate information from the internal knowledge base on courses, registration, finance, and certification to generate responses.
3.
We continuously refined this model and updated the knowledge base to reflect new information and policy changes.
Impact
Enhanced student support
Provided 24/7 accurate responses and reduced wait times.
Increased administrative efficiency
Reduced workload for administration and tutoring teams.
Tracked application usage
Monitored real-time application usage and engagement.