Hi, my name is
David Mgbede.
I'm a Software Engineer dedicated to building scalable web applications, AI-powered systems, SaaS products, and developer tools, designing robust backend services and engineering intelligent agentic workflows.
A bit about me
My journey as a software engineer has always been fueled by curiosity and a relentless drive to keep learning. I thrive on challenges that push me to think critically and creatively, whether it's solving a complex engineering problem, experimenting with emerging tech, or building new tools. I believe that every challenge is an opportunity to grow, and I approach them with patience, focus, and a commitment to continuous improvement.
This love for problem-solving is why I am passionate about building impactful, real-world products. Over the years, I have engineered web applications, AI-powered solutions, developer tools, and SaaS platforms using modern technologies like Next.js, TypeScript, Node.js, Python, and AI frameworks. I find it incredibly rewarding to turn raw ideas into finished products that people can actually use, while constantly refining how I design scalable and maintainable software.
Beyond the code, I believe that technology is ultimately about people. I find great fulfillment in giving back to the tech community,whether by mentoring aspiring developers, contributing to open-source, or sharing my experiences. I'm also deeply energized by collaborating with diverse minds. My ultimate goal is to bring enthusiasm and resilience to every team I join, engineering systems that improve lives while inspiring others to pursue excellence.
Technologies I work with
A collection of tools, frameworks, and languages I use to build modern, performant web experiences.
+ always learning more
Where I've worked
A selection of roles and companies I've contributed to professionally.
Contribute to the development of AI-powered products and intelligent software solutions for enterprise and business use cases. Design and implement LLM-powered features, agentic workflows, and scalable AI architectures utilizing retrieval, orchestration, and workflow automation components.
- Python
- FastAPI
- TypeScript
- React
- Next.js
- LLMs
- RAG
- Agentic Workflows
Contributed to the development of a hospitality management platform enabling restaurants and hotels to digitize operations, currently used by 200+ restaurants.
- Next.js
- TypeScript
- Tailwind
- ShadCN
- Zustand
- PostgreSQL
- Drizzle ORM
Led the frontend development for a comprehensive business management SaaS platform, helping fashion designers and cobblers digitize operations, manage customers, and accept payments, all in one place.
- Next.js
- TypeScript
- Tailwind
- ShadCN
- Zustand
- OpenAI
A creative media studio that tells stories through photography and videography. Captures moments, brings visions to life, and delivers high-quality media reflecting each client's unique identity.
- Next.js
- TypeScript
- Tailwind
- ShadCN
- GSAP
- Zustand
Led frontend development for V1 and V2 of a logistics and delivery platform serving merchants and riders across multiple regions. Built scalable, responsive user interfaces, improving Core Web Vitals scores and page load times.
- Next.js
- TypeScript
- Bootstrap
- SCSS
- HTML
Things I've built
A curated selection of personal and professional projects, each one solving a real problem.
Datrixs
Built an AI-powered data analysis platform that enables users to query structured and unstructured datasets using natural language. Developed a Python-based data processing service with FastAPI, pandas, OCR, and PDF parsing to ingest and analyze CSV, XLSX, PDF, and image-based data sources. Implemented Retrieval-Augmented Generation (RAG) pipelines using LanceDB and sentence-transformer embeddings to provide context-aware responses across uploaded datasets.
- Python
- FastAPI
- Mastra
- OpenRouter
- LanceDB
- Next.js
- TypeScript
- Prisma
- Pandas
Gloflix
Gloflix is a smart movie streaming platform powered by an AI recommendation system that suggests what to watch based on your mood, helping you skip the endless scrolling and find the right movie instantly.
- Next.js
- TypeScript
- Gemini
- Prisma
- NextAuth
- ShadCN
- Tailwind
- Zustand
Gluk
Gluk is an AI-powered research agent that helps you find, understand, and organize information faster—turning complex topics into clear, actionable insights so you can focus on building and learning. Added voice interaction capabilities using Deepgram's speech-to-text and text-to-speech APIs, enabling users to speak queries naturally and hear responses.
- Next.js
- Mastra
- TypeScript
- PostgreSQL
- Tailwind
- ShadCN
- Zustand
- OpenRouter
- Tavily
- Deepgram
AgriAgent
Built a production-grade multi-agent system using LangGraph, LangChain, and MCP architecture to coordinate intelligent workflows across specialized agents and external tools. Developed a FastMCP server and agent client capable of tool execution, knowledge retrieval, web search integration, and context-aware reasoning. Implemented hierarchical memory persistence using AsyncSqliteStore and vector embeddings.
- Python
- LangGraph
- LangChain
- FastMCP
- Neo4j
- Streamlit
- SQLite
- MCP
- Tavily
Forjex
A powerful command-line tool that eliminates repetitive setup tasks. Creates GitHub repos, writes CI/CD pipelines, crafts commit messages, sets up deployments, and initializes new projects.
- Node.js
- TypeScript
- Supabase
- Next.js
- Tailwind
- ShadCN
Olenar
Designed and built a learning management and mentorship platform using Next.js, TypeScript, Tailwind, PostgreSQL, and Supabase. Implemented authenticated dashboards, course enrollment, progress tracking, and mentor-learner interactions. Designed scalable frontend architecture that reduced component duplication by 30% and improved feature delivery speed.
- Next.js
- TypeScript
- Tailwind
- PostgreSQL
- Supabase
Community & Leadership
Where it all started, and where I still show up.
KodeCamp
KodeCamp is where my journey in tech officially began. Starting as a bootcamp student in 2022, I built the foundation for my software engineering career. I've since had the privilege of giving back to the community in various leadership roles, serving as Community Manager to facilitate growth and collaboration among 100+ developers, and currently as an AI for Developers Mentor in KodeCamp 6.0, guiding aspiring engineers in building LLM-powered applications and agentic workflows.
Currently serving as a Community Mentor for the AI for Developers track at KodeCamp 6.0, guiding students through modern AI engineering concepts including agentic workflows, RAG, MCP integrations, and AI application architecture.
Mentor aspiring developers in building AI-powered applications, reviewing projects, and providing technical feedback on Python and modern AI frameworks.
Served as Community Manager for 4 months, facilitating knowledge sharing, peer programming sessions, and structured discussions among 100+ developers.
Organized mentorship initiatives connecting junior developers with experienced engineers, helping members build practical skills and real portfolio projects.
Active community member, contributing learning resources and paying forward what the community gave me.
Thoughts & Writing
I write about AI engineering, LLMs, RAG, and the concepts most tutorials skip — helping 1,000+ developers understand what's actually happening under the hood.
Why Naive RAG Fails on Complex Data (And Nobody Tells You)
Most tutorials show you how to build a RAG pipeline. Almost none of them tell you where it quietly breaks — and why complex data exposes every weakness in a naive setup.
Read on MediumGet in touch
Have a project in mind, a question, or just want to say hi? My inbox is always open, I'll get back to you as soon as possible.