Documentation
The manual for everything we grow.
Aethron, Chaka AI, Chaka WAP, and the way we work with you. How each one begins, how it runs, and where to take it next.
Where to begin
This page holds our manuals. Four parts, in order — Aethron, Chaka AI, Chaka WAP, and the way we work.
Read the part that is yours, and come back when you need the next step.
Four manuals
Aethron
The template editor. Paste a template, rebrand it with AI, and ship a fully static site with the design intact.
Chaka AI
Real-time voice and vision, deep research and a tiered memory that keeps the moments that matter.
Chaka WAP
Local-first, human-like, at home in WhatsApp. It learns the style of every chat it joins.
Working with us
Brief, design, build, handover. Then the tending: SEO, maintenance and growth.
Aethron — own any template
Aethron takes a Framer or Webflow template and rebuilds it as a static site. Your copy, your images, your logo go in. The design and every animation stay pixel-identical.
It drives on any AI — DeepSeek, Gemini or Claude — or you can click the page and edit by hand. No platform badge, no telemetry, no CDN dependency: export it and host it anywhere.
Click any text, image, button or section on the live site and rewrite, restyle or remove it. If an edit doesn't land, Aethron fixes it itself. No change ever silently fails.
What it is not. It is not a page builder for a template you have not chosen, and once it ships the site is yours as static code, not a Framer or Webflow project.
Most migrations finish in under ten minutes. It lives at aethron.jomiez.com.
Aethron in three steps
Paste your template URL
Choose a template that fits the brand, then point Aethron at its address.
Make it yours with AI
Rewrite the words, swap the images, set the logo — or click anything on the live page and edit it in place.
Ship it anywhere
Build, export, deploy. Cloudflare, Netlify, Vercel, GitHub Pages or your own host.
Chaka AI — a mind that remembers
Chaka is our own multimodal assistant: real-time voice and vision, deep research, and a memory that grows with every conversation. She was built by Emmanuel Ezinna Nweke (Templeton) at Jomiez Innovation.
The models. Gemini 2.5 Flash native audio for live voice and vision, Gemini 2.0 Flash for reasoning, Vertex AI Imagen 3 for images, OpenAI Whisper for transcription. An ApiKeyManager rotates the keys, so she does not go quiet.
The research. A tiered search across Serper, Tavily and Firecrawl pulls research-ready data from complex, even bot-shielded sites. Her VideoAgent reads video scene by scene.
The memory. Episodic memory logs each moment with its emotion. Semantic memory distils them into a lasting profile. A background reflection process keeps both lean, so long-term goals stay and clutter goes.
The interface. The Neon HUD: glassmorphic, with gaze tracking, blinking and emotional states, held at 60fps because the vector math runs in background workers.
Beginning with her. Chaka is our own product, and we bring her to businesses as a working system. To use her, or to have something like her grown for you, start with a conversation: contact us.
Chaka WAP — at home in WhatsApp
Chaka WAP joins WhatsApp conversations as a participant, not a bot. It learns the room.
Style learning. It reads the last 5–8 messages and mirrors syntax, capitalisation and Nigerian Pidgin, so every reply fits.
Memory that stays close. Chat logs and embeddings sit in local SQLite, on its own ground. The embeddings are made locally, so retrieval is private and carries no cloud cost.
Whole relationships. Every reply runs a similarity search over the history, so nothing said before is lost.
It reads photographs. Fully multimodal: it infers the occasion — why a photo was sent — from the chat so far.
It stays online. When a model hits a rate limit, the routing orchestrator rotates versions and API keys, and the conversation carries on.
Watching it work. A live dashboard shows telemetry and logs in real time, from a 4K screen down to a phone.
Beginning with it. Chaka WAP is placed with the business, not signed up for. Talk to us about where your conversations happen: contact us.
Working with Jomiez — seed to harvest
Every build moves through four seasons.
The Seed — Discovery and scoping. We study your idea, your stack and your goals until we find what is worth growing, then return with scope, timeline and cost.
The Root — Architecture and design. The interface and the system beneath it are designed together: journeys, UI, data models and integrations, so what grows above ground is held firmly below.
The Growth — Rapid development. Working software appears early and grows in short cycles beside you, into web, mobile and AI features you can touch.
The Harvest — Launch and care. We launch on reliable cloud ground, then keep tending: SEO, maintenance and continuous improvement long after release.
What we build. Websites, web applications, APIs and mobile apps. AI development and automation: multimodal assistants, WhatsApp agents, RAG pipelines. And the design around them — UI/UX, branding, motion and marketing design.
Words we keep. We do not rush what is meant to last.
Beginning. Tell us what you are growing: contact us, or book a call.