Amariah Kamau
@amariahabish
AI Engineer building agentic RAG systems and production desktop AI platforms.
What I'm looking for
I’m an AI Engineer and Full Stack Architect who builds production-ready agentic AI systems—especially RAG pipelines and multi-agent orchestration—built to perform under real constraints.
I’m the Founder of Atlarix (v7), a native desktop AI coding environment shipped across Mac and Linux with full CI/CD, Apple Notarisation, and hardened release workflows. Atlarix uses a persistent Blueprint graph (SQLite) for RAG on every query, reducing token usage from ~100K to ~5K per query.
In Atlarix, I architected hierarchical context retrieval (Blueprint+Tree-sitter with BM25-scored targeting and Fast-model compression at 70% capacity) and a parallel agent setup (Research, Architect, Builder, Reviewer, Debugger) with task ledger continuity across context resets. I also shipped a managed AI tier that removes the need for user API keys, with monthly token limits enforced via Supabase RPCs.
Previously, I worked as an AI Specialist and Full Stack Engineer for scalable product integrations, including CMS + AI integration (Django REST, Next.js) and ML-driven business logic with real-time inference. I continue to focus on shipping reliable AI systems end-to-end—from model orchestration and infrastructure to distribution and monitoring.
Experience
Work history, roles, and key accomplishments
Co-founder & AI Engineer
NorahLabs
Nov 2025 - Present (6 months)
Sole engineer leading Atlarix (v1–v7), shipping a native desktop AI coding environment with blueprint-based RAG that reduced LLM tokens from ~100K to ~5K/query and autonomous multi-agent orchestration. Designed monetisation tiers (Free/Pro $19/mo/Workforce $79/mo) using LemonSqueezy webhooks with Supabase RLS and Auth0 Token Vault.
AI Specialist & Full Stack
Independent Consultant
Aug 2024 - Nov 2025 (1 year 3 months)
Architected an AI-integrated CMS with Django REST and Next.js, improving frontend performance by 20% using WebSocket-based real-time updates. Designed and deployed predictive ML models for complex business logic, including data preprocessing pipelines for real-time inference.
Software Engineer
Mizizi AI
Jun 2024 - Apr 2025 (10 months)
Built AI-powered fraud detection systems that contributed to a 15% improvement in conversion rates and integrated predictive analytics into the core product. Developed and maintained scalable RESTful APIs for 5,000+ monthly users with high availability and secure data handling.
Education
Degrees, certifications, and relevant coursework
Moringa School
Software Development Certification, Software Development
Completed a software development certification at Moringa School in 2024 and earned AWS Certified Data Engineer (Associate), Google Cloud Professional Data Engineer, and AWS Certified Developer.
Tech stack
Software and tools used professionally
GitHub
Docker Compose
GitHub Actions
PostgreSQL
SQLite
Gmail
Node.js
Django
Next.js
Tailwind CSS
Redis
Jira
React
JavaScript
Python
HTML5
CSS 3
Lua
FastAPI
Grafana
Prometheus
Linux
macOS
Gemini
Elasticsearch
TypeScript
Auth0
Docker
NGINX
SQL
Hugging Face
PostHog
Supabase
LangChain
Ollama
Ray
OpenAI API
Groq
Mermaid
Together AI
Bash
React Flow
Agentic
OpenRouter
LM Studio
Stack AI
Task
X++
Availability
Location
Authorized to work in
Portfolio
portfolio-pied-five-61.vercel.appSocial media
Job categories
Skills
Interested in hiring Amariah?
You can contact Amariah and 90k+ other talented remote workers on Himalayas.
Message AmariahFind your dream job
Sign up now and join over 100,000 remote workers who receive personalized job alerts, curated job matches, and more for free!
