Madu C.

I build production-ready web applications, AI-powered automations, computer vision pipelines, and scraping systems that help businesses reduce manual work, improve accuracy, and scale operations. My work combines full-stack development, backend systems, automation, data extraction, and applied AI. I help clients turn messy workflows into reliable software, whether that means building a React/Next.js frontend, creating a Python automation pipeline, training a computer vision model, or connecting everything into a production-ready backend workflow. I’m especially strong at building practical systems that solve real business problems, not just prototypes. I work across scoping, implementation, testing, refinement, documentation, and handoff. What I do best Full-Stack Development * React.js, Next.js, JavaScript, TypeScript * Node.js, Express.js * Responsive frontend implementation * API integrations and backend workflows * Admin dashboards, portals, and internal tools AI Automation & Applied AI * AI workflow automation * Prompt engineering and structured AI outputs * Retrieval-based Q&A systems * Model evaluation and AI testing * Internal tools powered by AI, APIs, triggers, and business logic Computer Vision & Model Training * Training and improving computer vision models for real-world use cases * Segmentation model refinement for cleaner, tighter, more accurate masks * Image recognition pipelines using image + text features * Model validation using metrics such as IoU, Dice score, pixel accuracy, and qualitative comparison workflows * Experience working with frameworks and models such as YOLO, SAM, Mask2Former, OneFormer, PyTorch, and related CV tooling Web Scraping & Data Extraction * Playwright, Puppeteer, Selenium, BeautifulSoup, Scrapy * Static and dynamic site scraping * Data extraction from websites, documents, emails, images, and PDFs * Data normalization, deduplication, and export workflows * Research automation and business data pipelines Cloud, Databases & Infrastructure * MongoDB, PostgreSQL, MySQL * AWS, Google Cloud, Firebase * Railway, Docker, Vercel * Deployment-ready backend services and automation workflows Selected Work 1. React/Next.js Frontend Engineering: Delivered a high-quality mobile-first frontend from design specs with pixel-perfect implementation, reusable components, and API integration. 2. Segmentation Model Training: Worked on improving segmentation quality for semantic regions such as walls, floors, ceilings, windows, furniture, and plants, with emphasis on smooth, tight, accurate masks and clean contour separation. 3. Image Recognition / Valuation Model: Worked on an AI system for identifying resellable couches, filtering low-value items, estimating value from image and text data, and detecting repeated-image scams for listing workflows. 4. Quotes and Reviews Scraping Pipelines: Built scraping systems for structured extraction across news and review-based datasets. 5. Python Automation & Data Extraction: Built automation and extraction systems for business workflows, structured processing, and operational efficiency. 6. Document-Based AI Workflows: Worked on retrieval-focused document Q&A and structured AI processing systems. I care about building systems that are clean, useful, and maintainable. I communicate clearly, move quickly, and structure projects so clients get reliable delivery with minimal friction. If you need someone who can build the software, automate the workflow, train or improve the model, and connect everything into a working system, I’m ready to help.