James Kinyua

Crafting High-Performance Distributed Systems & Conversational AI

I am an experienced Software Engineer passionate about building robust, user-centric software that solves real-world problems. While I specialize in scalable distributed backends and modern AI-enhanced applications, my primary focus is always on delivering reliable, production-ready value through rigorous testing, CI/CD, and comprehensive observability.

Explore My Work View Resume

About Me

I have a deep passion for software engineering and the incredible things accomplished by professional teams dedicated to this craft. From building simple web applications to software systems that power spacecraft, I am always eager to join teams that prioritize engineering excellence and reliable delivery.

Over the last decade, I have focused on mastering clean architecture, microservices, and AI integrations. Whether it's implementing asynchronous messaging architectures (RabbitMQ/NATS), or designing Retrieval-Augmented Generation (RAG) models, I go the extra mile to ensure my code is production-ready. I deeply embrace rigorous testing, CI/CD pipelines, and comprehensive observability to maintain system health and performance.

When I'm not at my keyboard architecting systems, you can usually find me cycling outdoors on my mountain bike or taking peaceful evening walks with my partner. I believe in maintaining a healthy, active balance to keep my mind sharp and creative.

James Kinyua

Projects Portfolio

Explore key projects spanning LLM integrations, scalable distributed backends, and responsive frontend applications.

Authentication & Session Isolation: For the optimal experience across the current Astra web apps, log in to one application at a time. To switch between these applications, log out of the current one first. This is because a custom shared domain has not been set up across the three deployments, and Clerk.js satellite domain configurations (to share sessions across distinct domains) are a paid Clerk Pro feature.

AI & RAG

Astra Document Summary

Next.js-based conversational AI web assistant designed for document parsing and summarization. Supports PDF, DOCX, or raw text uploads, providing cohesive summaries powered by Google Gemini.

Next.js 15 Vertex AI Vercel AI SDK Clerk Upstash Redis

Live Demo

AI & RAG

Astra Aviation RAG

Conversational AI safety assistant designed to contextually search and analyze NTSB aviation accident/incident reports using Retrieval-Augmented Generation and local Hierarchical Navigable Small World (HNSW) vector indexing.

Next.js RAG HNSWLib NestJS Vertex AI

Live Demo

AI & RAG

Astra Interview Assistant

Conversational AI application simulating real-world job interviews, asking tailored technical questions based on resumes and job descriptions, and delivering structured evaluations.

Next.js Vertex AI Vercel AI SDK TypeScript

Live Demo

Distributed/Cloud

ShiftSync

A full-stack, web-based shift scheduling platform designed for multi-location, multi-timezone restaurants. Resolves timezone translation issues and optimizes shift coverage.

Node.js React MongoDB Docker Timezones

Source Code

Distributed/Cloud

FlixTube

Cloud-native video streaming microservices platform. Features decoupled components for video processing and notifications utilizing event-driven pipelines on Kubernetes.

Docker Kubernetes Microservices RabbitMQ Node.js

Source Code

Technical Expertise

Core technology stacks, tooling, and architectural concepts acquired over years of development.

AI Workflows

LangChain Retrieval-Augmented Gen (RAG) Vercel AI SDK Google Vertex AI Gemini LLM HNSWLib

Languages

TypeScript JavaScript Python SQL Bash HTML5 / CSS3

Frontend

Next.js 15+ React React Native Tailwind CSS Redux Sass

Backend & APIs

Node.js NestJS Express.js Microservices GraphQL WebSockets

Storage & DBs

PostgreSQL MySQL MongoDB CosmosDB Redis Amazon S3

DevOps & Cloud

Docker Kubernetes Terraform GitHub Actions Microsoft Azure Google Cloud (GCP)

Education

Bachelor of Science in Information Technology

Mount Kenya University • Thika, Kenya

2012 - 2016 Research & Innovation Students Association

Let's Keep in Touch!

I'm always open to discussing new opportunities, open-source projects, or technical collaborations. Drop me a line!

LinkedIn Profile

Or explore my work on GitHub