Yohans (John) Bekele

Built and maintained cloud infrastructure using Pulumi and AWS across ECS Fargate, ALB, CloudFront, ACM, VPC, S3, SSM, IAM, and Route 53 — codifying multi-environment deployments for dev, QA, production, lab, and feature stacks.

Designed and operated advanced GitHub Actions CI/CD pipelines for build, test, deploy, regression, warmup, teardown, and release workflows — using reusable jobs, environment-scoped secrets, OIDC-based AWS access, and safe rollback patterns.

Developed AI-powered CI/CD workflows that automated PR review, infrastructure review, frontend-impact analysis, dependency remediation, and ticket-to-PR delivery — improving review quality while reducing repetitive engineering toil.

Implemented infrastructure guardrails for IaC safety, cross-stack consistency, secret handling, deployment readiness, and environment isolation — helping prevent misconfigurations before they reached production.

Automated operational workflows in Python, Bash, GitHub Actions, and monorepo tooling — streamlining deployments, health checks, log collection, release tasks, and recurring maintenance across multiple services.

Supported containerized application delivery on AWS ECS, Docker, ECR, ALB, and CloudFront — improving service reliability through health checks, structured deployment workflows, origin verification, and production warmup routines.

Strengthened observability and incident response through CloudWatch logs, ECS diagnostics, structured logging, PII-safe redaction, and AI-assisted SRE tooling — making production issues easier to detect, investigate, and resolve.

Authored and maintained technical documentation, runbooks, workflow standards, and AI agent instructions — establishing repeatable engineering practices for infrastructure, CI/CD, security, and automated delivery.

Built an AI-powered Knowledge Base management system using RAG and agentic AI, resolved 10,000+ production incidents, and achieved 99.8% system uptime through proactive monitoring and automation.

• Built an AI-powered Knowledge Base management system using RAG and agentic AI with agents that extract and synthesize information from multiple internal sources.
• Exposed the knowledge base through a robust REST API designed for integration with legacy applications, enabling end users to self-diagnose issues before opening tickets.
• Resolved 10,000+ production incidents via ServiceNow and JIRA, with structured root cause analysis, post-incident reviews, and runbook authoring.
• Fed recurring incident patterns back into the knowledge base to improve support effectiveness.
• Detected outages and client-side application faults early using clustering-based anomaly detection over user activity telemetry.
• Achieved 99.8% system uptime through proactive monitoring, alert tuning, and preventive maintenance across enterprise applications.
• Automated daily monitoring, health checks, and operational workflows in Python, Bash, and SQL, reducing manual effort by 15+ hours per week.
• Supported database operations and migrations across MS SQL Server, PostgreSQL, and MySQL.

Monitored application performance and infrastructure health to ensure high availability and SLA compliance across distributed systems.

• Tracked metrics, logs, events, and traces using New Relic and Datadog across distributed production environments.
• Managed incident triage, event correlation, and escalation via ServiceNow and JIRA with signal enrichment and actionable alerting.
• Reduced mean time to resolution (MTTR) by 45% through improved incident management workflows.
• Automated routine operational tasks, health checks, and reporting with Python, Bash, and SQL scripts, improving team efficiency by 30%.
• Built custom dashboards for anomaly detection and proactive alerting to surface issues before customer impact.
• Supported database migrations and integrity verification across MS SQL Server, PostgreSQL, and MySQL production environments.
• Created runbooks and automated remediation scripts for incident response procedures, enabling faster recovery and reduced human intervention.
• Collaborated with development and infrastructure teams to implement observability solutions, reduce alert noise, and improve system reliability.

Built end-to-end AI and web applications across finance, education, legal, real estate, agriculture, and infrastructure automation domains.

• Developed LLM-powered features including financial news analysis, entity extraction, sentiment analysis, stock-signal generation, and AI tutoring workflows with prompt design and evaluation.
• Designed full-stack architectures with REST APIs, authentication flows, database schemas, background processing, file handling, user dashboards, admin panels, and responsive frontends.
• Built infrastructure automation tooling translating visual input into Pulumi/Terraform code, with drag-and-drop AWS modeling, GitHub OAuth integration, and repository workflows.
• Implemented production engineering practices including Docker containerization, API testing, reusable components, service separation, environment configuration, CI/CD setup, and deployment troubleshooting.
• Used React, Next.js, TypeScript, Python/FastAPI, Node.js/Express, PostgreSQL, MongoDB, Docker, and cloud deployment workflows across all projects.

Executed large-scale data migration and transformation projects across multiple source systems, improving data processing efficiency by 30%.

• Automated ETL pipelines using Python, Bash, and SQL to streamline migration workflows.
• Designed data validation and cleansing routines to ensure accuracy, consistency, and compliance during migrations.
• Built data ingestion pipelines that normalized and enriched datasets from heterogeneous sources.
• Created dashboards and analytical reports providing actionable insights on migrated datasets for business continuity and compliance.
• Authored technical documentation and best practices guides for migration workflows and automation scripts.
• Collaborated with cross-functional teams to ensure seamless data integration with minimal disruption to business operations.

Built and maintained secure, responsive React-based banking web interfaces across the MERN stack, covering component architecture, hooks-based state management, REST API integration, form validation, and production-ready layouts.

• Developed internal banking operations dashboards and admin portals used by business and operations teams to monitor transactions, manage customer and account records, generate reports, and support daily reconciliation workflows.
• Delivered customer-facing digital banking pages including product information, service pages, onboarding flows, and transactional web surfaces with a focus on accessibility, mobile responsiveness, browser compatibility, and clear user journeys.
• Partnered with backend engineers on Node.js/Express API contracts and MongoDB schemas, ensuring accurate data handling for customer, transaction, reporting, and operational workflows.
• Implemented frontend authentication and role-based access controls for internal banking tools, including protected routes, session-aware UI behavior, and conditional access to sensitive operational screens.
• Maintained reusable UI components, styling conventions, and shared frontend patterns across internal and public-facing banking applications, improving consistency, reducing duplication, and accelerating delivery of new screens.
• Improved frontend performance through code splitting, lazy loading, asset optimization, and bundle-size control, helping banking web services remain usable on slower networks and lower-end devices.
• Supported production quality through component testing, linting, peer code reviews, defect fixes, and regression checks, helping reduce frontend issues before they reached banking staff or customers.