AI · Conversational · 2024

IrenicBot

Conversational AI that remembers.

A conversational AI assistant built on Gemini — text and image input, saved conversations, favourites, and an admin surface for usage analytics.

IrenicBot Gemini-powered conversational AI app: chat screens on iPhone
Shipped in 2024
IrenicBot Gemini conversational AI: composite product mockup
IrenicBot Gemini conversational AI: chat view with saved conversation history
IrenicBot Gemini conversational AI: multimodal text and image input
IrenicBot Gemini conversational AI: saved conversations and favourites

Gemini

Reasoning core

Text + image

Inputs

2024

Shipped

Conversational AI that remembers.

The story

How it came together.

Problem

Off-the-shelf chatbots either look generic or lock customers into one vendor. The product layer — saved threads, favourites, multimodal input, account history — is where the value lives.

Approach

Gemini for the reasoning core, Firebase for accounts and conversation state, Flutter for a single codebase across iOS and Android. Designed admin analytics so operators can see how the bot is actually being used.

Outcome

A conversational assistant that handles text and images, remembers context per user, and exposes meaningful product analytics to operators.

What people said

Product Reviews.

Honest takes from people who actually use the product.

Carlos F.

E-commerce shop owner

Answers most product questions before they reach my inbox. The image input means a customer can drop a screenshot of a label and get a direct answer.

Reema S.

Independent researcher

We use it as a research assistant. The conversation history survives between sessions, which is the thing that breaks every other AI tool we've tried.

Khalid M.

Translator

Image-plus-text input is the killer feature. I send a photo of foreign signage, ask a question, get a useful answer. Saves real time.

Niamh O.

Marketing lead

We set it up as a customer-facing chatbot on our SaaS marketing site. The admin analytics let us see what people actually ask, which has reshaped our help docs.

Vikram D.

Indie developer

Multimodal AI without the enterprise pricing. Saved conversations and favourites turn it from a toy into a tool.

More work

Keep exploring.

All projects

Start a conversation

Tell us what you’re building.

Share the essentials and we’ll reply within 4 hours with a real next step, not an auto-responder.

What happens next

  1. We reply within 4 hours, from a real person, not an auto-responder.
  2. A short scoping call to understand the goal, constraints, and timeline.
  3. A fixed-scope discovery sprint: a working prototype and a written estimate.
Office
Austin, TX, United States
Hours
Mon–Fri · Async + scheduled calls

Fields marked are required.