Senior Software Engineer with 10+ years delivering scalable enterprise applications across fintech, banking, and cloud-native environments. I combine deep software engineering fundamentals with modern AI capabilities to ship systems that are intelligent, reliable, and production-ready β not just prototypes.
What I build:
- AI-powered applications with LLMs, RAG pipelines, and agentic tool-use
- Full-stack products serving real users at scale (React/Next.js + FastAPI/Spring Boot)
- Serverless cloud infrastructure on AWS with Terraform and Docker
- Internal tooling and workflow automation that cuts manual effort
Recent impact:
- Contributed to the M36 Investment Platform at Union Bank of Nigeria, supporting 200%+ revenue growth through product feature delivery
- Built and deployed a production AI Digital Twin hosted on AWS β live at yaqub.online
- Developed agentic systems using LangGraph, Claude, and MCP that autonomously resolve engineering tasks end-to-end
| Area | Details |
|---|---|
| π€ Agentic AI | Multi-agent orchestration, autonomous workflows, LangGraph |
| π§ RAG Systems | Production retrieval pipelines, vector search, context-aware LLMs |
| π MCP Integration | Model Context Protocol tooling and server design |
| βοΈ Cloud Engineering | AWS Lambda, Terraform, serverless-first architecture |
| π AI-Native Frontend | Next.js applications wired to LLM backends |
Business problem: Engineering teams waste hours triaging and patching recurring JS/TS bugs manually.
An autonomous multi-agent workflow that identifies, analyzes, fixes, verifies, and opens pull requests for JavaScript and TypeScript repositories β without human intervention.
- Architecture: LangGraph agent loop β Claude Opus reasoning β GPT code synthesis β automated PR creation
- Deployment: FastAPI backend + Next.js dashboard + SSE streaming
- Auth: Clerk-secured multi-tenant access
Stack: Python Β· FastAPI Β· LangGraph Β· Claude Opus Β· Next.js Β· Clerk Β· SSE
Business problem: Professionals can't personally respond to every recruiter or collaborator inquiry β a trained AI persona can.
A personalized AI assistant that reasons and responds as me, deployed on production AWS infrastructure with Terraform-managed serverless architecture.
- Architecture: Next.js frontend β AWS Lambda β FastAPI β LLM orchestration layer
- Infrastructure: Fully IaC-managed with Terraform; zero-server deployment
- Live demo: yaqub.online
Stack: Python Β· FastAPI Β· Next.js Β· AWS Lambda Β· Terraform
Business problem: Customer support operations are bottlenecked by repetitive queries that don't need a human agent.
A production-grade AI support assistant with customer authentication, order lookup, product availability checks, and order placement β all driven by an agentic LLM tool-use loop over MCP.
- Architecture: Gradio UI β ChatEngine β OpenRouter LLM β MCP Tool Server (GCP)
- Capabilities: Auth, order history, inventory checks, order placement
- Deployment: HuggingFace Spaces + GCP MCP backend
Stack: Python Β· MCP Β· OpenRouter Β· Gradio Β· GCP
Business problem: Incident resolution requires coordination across multiple systems under pressure β an agent can triage faster than any on-call engineer.
An AI-driven multi-agent system for automated incident analysis, response coordination, and operational recovery β reducing mean time to resolution.
Stack: Python Β· Multi-Agent Systems Β· LLMs
Business problem: Agribusinesses lack affordable digital tools to manage farm operations, sales, and logistics.
Full-stack agricultural platform for operational management and digital workflows, built with enterprise-grade Spring Boot backend and Next.js frontend.
Stack: Java Β· Spring Boot Β· Next.js Β· TypeScript
I'm actively looking for my next role. If you're hiring or know someone who is, let's talk.
- Senior Software Engineer (AI applications focus)
- AI Application Engineer / Forward-Deployed Engineer
- Full-Stack Engineer (React/Next.js + Python/Java backend)
- Platform / Cloud Engineer
- Technical Consulting & Contract Work
π Lagos, Nigeria Β· Open to remote and relocation




