success

Statistics · MCP · Tools

success

I notice problems that don't have good solutions yet — then I build the tools. Currently focused on MCP servers that give AI assistants real, structured capabilities.

About

Builder. Explorer. Problem solver.

I'm a 200-level B.Sc Statistics student at Federal University Wukari, Nigeria. I combine a statistics foundation with real software development — which means I don't just write code, I understand the data behind it.

Most of what I build sits at that intersection: tools that pull live data, process it in a meaningful way, and surface something useful to the person on the other end.

Right now I'm focused on MCP (Model Context Protocol) servers — building tools that plug directly into AI assistants like Claude and Grok and give them real, structured capabilities.

I love exploring ideas and solving problems. When I notice something that doesn't have a good solution yet, I can't leave it alone — I start building.

Previously I moderated and managed communities for NFT projects. That experience taught me how to coordinate people, ship under pressure, and keep things running when everything is moving fast.

Everything I build is deployed, not just pushed to GitHub. Open to internships, collaborations, and entry-level opportunities in data science, analytics, or AI development.

Selected Work

Tools that actually ship

A selection of MCP servers and applications I've built. Each one started as an idea and ended as a working, deployed product.

FlowDev

MCP Server
Live

A remote MCP server that connects any MCP-compatible AI directly to a developer's VS Code workspace. Your AI reads files, writes code, searches your codebase, runs terminal commands, and pushes to GitHub — all from the chat window. No copy-pasting. No tab switching.

How it works

The MCP server acts as a relay. Your AI sends tool calls to FlowDev, which forwards them to a VS Code extension via persistent WebSocket. The extension executes locally and returns results in real time. Your code never leaves your machine.

TypeScriptNode.jsSocket.ioVS Code ExtensionNeon PostgreSQL

RiskLens MCP

MCP Server
Live

Risk analysis for US public companies using live SEC EDGAR data. Pulls Form 8-K filings and Form 4 insider transactions, classifies them against a built-in risk taxonomy, detects clustering patterns, and returns a scored risk assessment with a plain-language explanation.

How it works

Two focused tools: analyze_8k_events (material events, restatements, leadership changes, impairments) and analyze_insider_activity (clustered selling vs routine activity). Backed by Redis caching, rate-limited SEC calls, and designed for concurrent users.

PythonFastMCPRedisSEC EDGARBeautifulSoup

ClipNest

Full-stack App
Live

Free video downloader that supports TikTok, Instagram, Twitter, Facebook, Vimeo and 1000+ other platforms. Paste a link, choose format and quality, download instantly — no watermark, no signup required.

How it works

Next.js frontend with a clean queue-based interface. Paste multiple links, select format (video or audio), and download. Built for speed and simplicity.

TypeScriptNext.jsTailwind CSS

Approach

How I work

Notice

I look for problems that don't have good solutions yet — especially at the intersection of data, AI, and real workflows.

Explore

I dig into the data, the APIs, and the constraints. Ideas get tested quickly against reality.

Build

I own the full stack from idea to deployment. MCP servers, data pipelines, full apps — shipped, not just sketched.

Iterate

Everything is deployed early. Feedback from real use drives the next version.

Skills

What I work with

MCP & AI

Model Context ProtocolFastMCPAI AutomationTool DesignAgent Integrations

Data & Python

PythonData AnalysisSEC EDGARRedisBeautifulSoupPandas

Full-stack

TypeScriptNext.jsReactNode.jsExpressPostgreSQL

Community & Web3

Community ModerationCollab ManagementNFT ProjectsDiscordOn-chain

Exploration

Ideas & new tools

I constantly have ideas for new tools and technology. Some become MCP servers. Some become full applications. The common thread: they start from a real problem I noticed and couldn't ignore.

Currently exploring more tools at the intersection of AI and real financial / business data, deeper data science and machine learning applications, and ways to make AI assistants more capable through well-designed MCP servers.

Contact

Let's build something

Open to internships, collaborations, and interesting problems. If you're working on something that needs someone who can both understand the data and build around it, reach out.