This project demonstrates a transition from anecdotal learner feedback to a structured technical telemetry model. By architecting a 'Sidecar Ledger' in SQLite, I quantified architectural bottlenecks in technical training labs, identifying that 40% of all learner downtime originated from a single service domain (IAM).
Using methodologies refined during my experience at AWS, I treated the learning environment as a distributed system, logging error signatures to drive targeted engineering refactors.
To move beyond subjective feedback, I built a local relational database to capture high-fidelity student error logs. This allowed for the normalization of "blocker" tickets into a queryable format.
- Database: SQLite3
- Schema Design: Implemented Foreign Key constraints and Lookup Tables to ensure standardized categorization across AWS Service Domains (e.g., S3, IAM, Lambda).
- Normalization: Mapped disparate error strings (e.g., "Access Denied" vs. "403 Forbidden") to specific architectural failure modes.
Using SQL aggregation, I performed a deep-dive analysis of friction distribution across the lab architecture. The data revealed that while students often reported difficulty with code-heavy services (Boto3/Lambda), the primary bottleneck was actually upstream in the identity layer.
| Infrastructure Component | Total Logged Incidents | % of Total Friction |
|---|---|---|
| Identity & Access (IAM) | 64 | 40.0% |
| Storage Services (S3) | 32 | 20.0% |
| Lambda / Boto3 | 42 | 26.6% |
| EC2 Compute | 11 | 6.7% |
| CloudWatch / Logs | 11 | 6.7% |
By isolating the IAM domain, the data highlighted a specific "Critical Path" failure:
- Primary Failure Mode:
iam:PassRoleandPrincipalpolicy misconfigurations. - The Magnitude: This single failure type accounted for 83% of all domain-specific issues, causing a 30% decrease in student lab velocity.
Based on the data, I implemented two targeted interventions to neutralize the identified friction points:
- Automated Pre-Flight Validation: Developed a bash-based validator script that checks for the existence of required IAM Roles and permissions via the AWS CLI before a student attempts deployment.
- Documentation Refactor: Reallocated 50% of technical writing cycles to the Identity Management module, introducing visual policy-logic flowcharts to clarify complex permission inheritance.
- Measurable Outcome: These system refactors are projected to increase student throughput by 25% for future cohorts.
/database/error_telemetry.db: SQLite database containing normalized, anonymized friction logs./sql/schema.sql: DDL scripts including table constraints and relational mapping./sql/bottleneck_analysis.sql: SQL queries used for frequency analysis and impact cross-tabulations.