class SoftwareEngineer:
    def __init__(self):
        self.name = "Herbert Kwame Yeboah"
        self.role = "Software Engineer & AI/ML Developer"
        self.location = "Ghana 🇬🇭"
        self.education = "BSc Computer Science - UENR (2024)"
        self.current_focus = "AI-powered Solutions for Real-World Impact"
        
    def get_skills(self):
        return {
            "languages": ["Python", "TypeScript", "JavaScript", "SQL", "C++", "Java"],
            "ai_ml": ["TensorFlow", "Keras", "Scikit-learn", "NumPy", "Pandas"],
            "web": ["Next.js", "React", "Node.js", "Flask", "Tailwind CSS"],
            "cloud": ["Google Cloud", "Firebase", "Docker", "Railway"],
            "databases": ["PostgreSQL", "MongoDB", "Redis", "Firestore"]
        }
    
    def get_certifications(self):
        return [
            "ALX Africa - AI Career Essentials (2024)",
            "ALX Africa - Software Engineering (2024)",
            "Udemy - SQL Masterclass (2024)"
        ]

me = SoftwareEngineer()

📍 Based in Ghana 🇬🇭 | 🎓 BSc Computer Science (2024)

I am a Software Engineer & AI/ML Developer passionate about transforming complex ideas into production-ready code. My foundation blends high-performance software engineering with expertise in Machine Learning, allowing me to build end-to-end intelligent systems that solve real-world problems and make life better through technology.

Zero-Cost, Crash-Proof LLM Orchestration Framework

NeurIPS 2025 research analyzing 1,642 real-world multi-agent execution traces found a 41–86.7% failure rate across 7 state-of-the-art open-source systems. The root cause: cascading error propagation. DagPipe makes cascade failure structurally impossible.

The reliability layer that makes AI workflows safe to ship — crash recovery, schema validation, and intelligent cost routing — in 150 lines of Python. Runs entirely on free-tier APIs. Zero infrastructure. Zero subscription.

Pipeline: research → outline → draft → edit → publish
                                  ↑
                            crashed here

Re-run → research ✓ (restored) → outline ✓ (restored) → draft (re-runs) → ...

Technical Highlights:

Key Features (v0.2.0):

• 🔁 Crash Recovery — JSON checkpointing per node; resume exactly where you stopped
• 🧠 Smart Model Router — auto-selects model by task complexity; escalates on failure/rate-limit
• 📋 Constrained Generation — Pydantic schema validation with auto-retry on malformed output
• 🔒 Context Isolation — nodes only access their declared dependencies; safe for sensitive data
• 🗂️ Live Model Registry — self-maintaining database of free-tier availability; refreshes every 24h
• ⚙️ Pluggable Checkpoint Backends — swap filesystem for Redis, S3, or any custom store
• 🌐 MCP Server — generate crash-proof pipelines via Claude Desktop, Cursor, or Windsurf

Test Coverage: 108 tests · 5 modules · 0 regressions · Python 3.12 + 3.13

Available On:

pip install dagpipe-core

Python       12 hrs 45 mins  ███████████░░░░░░░  45.2%
TypeScript   8 hrs 30 mins   ███████░░░░░░░░░░░  30.1%
JavaScript   3 hrs 15 mins   ███░░░░░░░░░░░░░░░  11.5%
SQL          2 hrs 10 mins   ██░░░░░░░░░░░░░░░░   7.7%
Other        1 hr 30 mins    █░░░░░░░░░░░░░░░░░   5.5%

💡 "Using AI to solve real-world problems and make life better"