I'm a fullstack engineer specializing in AI security and production-grade automation systems. I design and implement end-to-end solutions that prioritize reliability, security, and maintainability. My work focuses on building intelligent systems that can operate autonomously while maintaining human oversight and control.

Building secure, scalable AI systems with a focus on production reliability, human-in-the-loop workflows, and self-healing automation.

• AI security patterns including human-in-the-loop validation and self-healing systems
• Production automation tools with custom databases and AWS deployment
• Multi-tenant architecture and microservices design
• Test-driven development for critical AI workflows
• LLM agent development and orchestration
• RAG system implementation and optimization

Primary: Python, JavaScript, TypeScript
Backend: FastAPI, Node.js, Express
Frontend: React, Next.js
Database: PostgreSQL, MongoDB, Redis

• LLM agent development and orchestration
• RAG system implementation and optimization
• Evaluation frameworks and testing harnesses
• Automation tools: Make, n8n
• Custom automation platforms with owned databases

• Multi-tenant system architecture
• Microservices design and implementation
• Event-driven architectures
• API design and versioning
• Test-driven development (TDD)

• Human-in-the-loop AI workflows
• Self-healing system design
• Automated monitoring and alerting
• Security best practices for AI systems
• Error detection and recovery patterns

• AWS deployment and management
• Docker containerization
• CI/CD pipeline design
• Infrastructure as Code
• Production monitoring and logging

Production-ready AI security system with human oversight capabilities. Implements multi-layer validation, automated threat detection, and self-healing mechanisms for LLM-based applications.

Key Features:

• Human-in-the-loop approval workflows
• Real-time security monitoring
• Automated incident response
• Multi-tenant architecture
• Comprehensive audit logging

Tech Stack: Python, FastAPI, PostgreSQL, Redis, AWS

Self-hosted automation platform with visual workflow builder and custom database integration. Deployed on AWS with high availability and scalability.

Key Features:

• Visual workflow designer
• Custom database connectors
• Event-driven execution
• Webhook integrations
• Real-time monitoring dashboard

Tech Stack: TypeScript, Node.js, PostgreSQL, n8n, AWS ECS

Comprehensive testing and evaluation system for retrieval-augmented generation applications. Includes golden datasets, scoring rubrics, and regression detection.

Key Features:

• Automated test generation
• Multi-dimensional scoring
• Hallucination detection
• Performance regression tracking
• Continuous evaluation pipeline

Tech Stack: Python, LangChain, ChromaDB, MLflow

Architecture First: I prioritize system design and scalability from the start, focusing on patterns like microservices, multi-tenancy, and event-driven architectures.

Test-Driven Development: I write tests before implementation to ensure code quality, maintainability, and regression prevention.

Security by Design: Security considerations are integrated throughout the development process, not added as an afterthought.

Production Ready: Every system I build is designed for production from day one, with proper monitoring, logging, error handling, and recovery mechanisms.

Email: eshetieyabibal@gmail.com
LinkedIn: linkedin.com/in/yabibal-eshetie
GitHub: github.com/Yab112