Ease

Framing the Problem

What is the problem you are trying to solve? Who does it affect?

Smart homes are still out of reach for most families—not because the tech isn’t available, but because it can be fragmented, expensive, and hard to integrate. Household tasks such as laundry is repetitive but yet essential. It demands attention, timing, and decision-making.

While often manageable for some, it poses significant challenges for individuals with executive dysfunction, memory and sensory difficulties, or neurodivergent conditions such as ADHD or autism. Tasks like remembering to run loads, knowing which settings to use for different fabrics, rotating towels and bedsheets on time, and maintaining the machine (like cleaning filters or descaling) create a high cognitive load.

In addition to this, neurodiverse individuals often face barriers when interacting with technology such as sensory overload, difficulty processing complex information, or challenges with navigation and focus.

The current situation:

• Over 80% of washing machines, ovens, and fridges in UK households are not internet-connected or smart-enabled. (UK Office for National Statistics, 2023)

• The average household juggles over 60 recurring domestic tasks weekly. (British Journal of Psychology, 2022)

• Cognitive load and mental fatigue from 'invisible labour'—like remembering when the laundry’s done—is one of the top sources of household stress, especially for carers and parents. (Harvard Business Review, 2021)

• There is an estimated 15% of the UK population are neurodivergent. There are over 700000 adults with autism, 1.9 million with ADHD and and estimated 10% with diagnosed dyslexia. 1 in 6 people globally have a disability.

• Smart plugs and door sensors cost under £20 and are now widely available, making retrofitting more accessible and affordable than ever. No need to buy a smart appliance to make your life easier.

Idea Explanation - Solution

Ease is an app that uses an AI-powered home assistant (EaseAI) that works with cost-effective, retrofit sensors to help any individual household take control of their appliances, starting with laundry.

✅UX and UI:

It is designed with accessibility in mind as this enhances usability for everyone, with simple, distraction-free interface with visual cues, voice assistant options, and customisable routines (e.g., “Monday dark wash”)

✅ Core Features:

• Onboarding

  - Device specific functions:

  QR code scan for the manual, to determine device-specific requirements/routines/functions to train A.I.

  - User profile questions to customise user needs

  - Calendar syncing and smart scheduling:
  Schedule wash cycles in advance via app or voice, with calendar integration with multiple users (e.g., Google Calendar). Detects your schedule (eg. school/work days, sports, periods) for specific-type washing

• Chatbot prompted activities

  • Reminders & Notifications:
    Timely alerts to remove clothes post-wash, clean filters, or descale the machine.

    If connected with an inbuilt smart device, it is able to turn on the washing machine at scheduled times etc (IoT actuation)

  • Replenish laundry consumables

  • Memory (eg. "I noticed you are away every last weekend of the month - shall I skip laundry that day from now on?")

• Energy Cost Optimization:
  Connects with apps like Octopus Energy to identify off-peak times for washing—starts cycles when electricity is cheapest.

Implementation

How do all the pieces fit together? Does your frontend make requests to your backend? Where does your database fit in?

Our solution includes:

• A frontend app (React) where users can:

  • Add household members and shared laundry needs

  • Set up recurring reminders (e.g., wash towels every Friday)

  • Mark tasks as done (e.g., “sheets washed”, “filter cleaned”)

  • Receive real-time nudges or notifications when laundry stages are delayed

• A lightweight backend (Node.js + Express or Firebase) that:

  • Stores household laundry schedules and task logs

  • Sends reminders at the right time (via push notifications or in-app alerts)

  • Learns task frequency patterns to suggest future actions

• Calendar & Energy Sync

  • The app fetches calendar events via the Google Calendar API

  • It also checks energy pricing data from the Octopus Energy API (or user predefined rates)

  • A simple rule-based logic engine finds the best time to wash based on:

    • User availability

    • Electricity cost

    • Clothing urgency

• A simple database (e.g. mySQL) tracks

  • Task history (when towels were last washed), preferred times, schedules

  • Per-user preferences for notification styles and formats

• AI/Machine Learning Integration and automation with user data collected

• Use machine learning modelling (e.g., Decision Trees, clustering, Neural Networks or reinforcement learning) that learns from user input, and feeds back with FastAPIs.

  • calendar events

  • user choices (when they accept or delay washes)

  • Adjusts recommendations over time
    

Challenges

What did you struggle with? How did you overcome it?

• Scoping for the hackathon
  The laundry journey involves many micro-steps. We focused on the most common and impactful ones first (e.g., reminders and task logging) to stay lean.

• Accessibility-first UX
  Designing for neurodivergent users requires multiple modes of interaction. We implemented text, icon, and color-based cues — but hope to expand to voice control and screen reader optimization next.

• User testing in a short timeframe
  We used lived experience from team members and anecdotal feedback from neurodiverse friends to shape feature priorities.

• Building with Machine learning, AI features

  Due to time constraints, user input data availability and limited expertise within the team to create fullstack with AI/ML models and integration. We decided however to describe the possible process, keeping feasibility and future build in mind, and provide a simulated view or sample code.

Accomplishments

What did you learn? What did you accomplish?

Next Steps

What are the next steps for your project? How can you improve it?

• Integrate GPT or other LLMs to make the assistant more conversational and adaptive. Learning from a more diverse user background will also reduce training bias of the AI models.

• Using Fabric Recognition Models to identify clothing type/material and recommend safe wash settings (e.g., cotton vs. silk).

  Leverage existing datasets (eg. TextileNet - [2301.06160] TextileNet: A Material Taxonomy-based Fashion Textile Dataset), using deep learning and Convoluted Neural Networks trained to recognise fabric types (e.g., cotton, wool, denim)

• EaseAI as an Agent

  • Understands routines without asking every time

  • Plans & acts on its own (with user permission)

  • Learns from behaviour and improves

  • Explains why it acts (“I ran the wash early to save £0.30 using off-peak rates.”)

  • Supports additional tasks (cleaning, meds, bins) with goal planning

Monetisation Potential

The app will follow a freemium model:

• Free core features for all users

• Premium subscription for advanced scheduling, energy optimisation, multi-device sync, and personalised AI coaching. Industry partners (eg. appliance brands, smart plugs) may wish to also offer this to their customers)

• Become an affiliate partner with online retailers.

Ease is designed for global scalability, starting with key markets in the UK, EU, and US where household task automation, smart home adoption, and awareness of neurodiversity are rapidly growing.

User Activity Assumptions

• 35–40% Monthly Active Users (MAUs)
  → 105,000 MAUs (based on benchmarks from apps like Notion, Todoist, Headspace)

• 10–12% conversion to paid subscription
  → 11% used here, based on high-utility, habit-forming freemium products (e.g., Calm 10–15%, Todoist ~8%)

Annualised Potential: ~£1.92M

As the product expands to cover additional household tasks, the per-household value and revenue per user are expected to increase accordingly.