Apps exclude elders
Elderly parents living at home cannot operate smart-device apps — menus, pairing, updates. The devices end up controlled only by the children who installed them.
Agentic AI for Indian homes · Bengaluru
A fully autonomous AI brain for Indian homes
No app to learn — the family simply talks to the home on WhatsApp, by voice, in Hindi, English, Kannada or Marathi.
Built first for elderly parents living at home, on the affordable smart devices Indian families already own.
Languages at launch
- हिन्दी Hindi
- English
- ಕನ್ನಡ Kannada
- मराठी Marathi
Voice notes on WhatsAppNothing new to install or read
Four languages at launchHindi, English, Kannada, Marathi
Brand-agnosticOne brain across popular smart-device brands
Offline-firstA local hub keeps routines running
One minute, one home
A day with the agents
From an afternoon voice note to the next morning’s electricity report: what the agents notice, what they decide, what they switch, and when they bring the family in — in the order it happens.
- 0:00Six apps, none of them usedThe whole home moves into one WhatsApp chat.
- 0:10A voice note at three o’clock“It’s too hot” — the agents read the room before doing anything.
- 0:20Plan, act, explainFan to high, curtains closed, and a reply in Marathi saying why.
- 0:30Nobody had to askAt eight the medicine shelf lights up and a gentle reminder follows.
- 0:40Family stays in the loopA tablet missed for twenty minutes reaches her son in Dubai.
- 0:50Night, then morningLights dim once she sleeps; the report arrives with units and rupees.
Read the narration
- Asha has six smart-home apps and uses none. So she got SmartHomeAI, a team of agents she talks to on WhatsApp.
- One afternoon she sends a voice note saying it’s too hot, and the agents first read the room: thirty-four degrees, three o’clock.
- They plan the way a caring son would, fan to high and curtains closed, then act and reply in Marathi explaining why.
- Most of their work happens unprompted: at eight o’clock the medicine shelf glows and a gentle voice note reminds Asha about her tablet.
- The twist is that the agents also send Rohan a WhatsApp alert in Dubai whenever Asha misses that tablet for twenty minutes.
- By night they dim the lights once she sleeps, and by morning they report every unit and rupee: six apps became one conversation.
The problem
Elders at home cannot use smart devices
Smart devices are cheap and everywhere in India. What is missing is the intelligence layer above them — one that is affordable, multilingual and trustworthy.
One app per brand
Every brand ships its own app. The family becomes the integration layer, and every new device adds another app.
English-first voice
Mainstream voice assistants handle Hindi only partially, and Kannada or Marathi poorly. Elders who speak only their mother tongue are left out.
Reactive, not autonomous
Automations are still “if motion, then light”. No goals, no planning, no explanation. The home reacts; it does not look after anyone.
From smart to agentic
What changes when a device has goals
A smart device reacts to a trigger. An agentic device pursues a goal — it observes, reasons, plans, acts and reflects.
| Capability | Smart device (today) | Agentic device (SmartHomeAI) |
|---|---|---|
| Follows rules | Yes | Yes — and knows when to override them |
| Has goals | No | Yes: comfort, energy budget, safety |
| Handles ambiguity | No | Reasons through it in natural language |
| Multi-step planning | No | Plans a day, not a single action |
| Explains decisions | No | Every action carries a why |
| Self-corrects | No | Observes outcomes and updates memory |
| Coordinates devices | Limited scenes | Multi-agent negotiation across the home |
How it thinks
The agent loop: observe, reason, plan, act, reflect
Follow one request through the loop. A 72-year-old sends a WhatsApp voice note in Marathi: “खूप गरम होत आहे” — “it is very hot.” She expresses a feeling, not a command; the system carries the reasoning.
Step 1 of 5
Observe
The agent gathers context: sensor readings, who is at home, the electricity tariff, the weather — and any request from the family.
In the example
A Marathi voice note arrives. The room sensor reads 34 °C and it is 3 pm.
Step 2 of 5
Reason
A language model and planner interpret the goal behind the request in its context — even when the request is a feeling rather than a command.
In the example
“It is very hot” is a comfort goal, not an instruction. The agent weighs the time, the temperature and her past preference.
Step 3 of 5
Plan
The goal becomes an ordered list of tool calls, each with an expected outcome that can be checked afterwards.
In the example
- Fan to high
- Curtains closed
- Confirm in Marathi, with the reason
Step 4 of 5
Act
The plan is carried out through the devices themselves, with notifications to the family where needed.
In the example
The fan and curtains are switched, and she receives a Marathi voice note confirming what was done and why.
Step 5 of 5
Reflect
Did it work? The agent checks the outcome against what it expected, then updates its memory and her preferences.
In the example
It confirms the devices responded and records the outcome, so the next hot afternoon starts from what she prefers.
No app to learn
Talk to the home on WhatsApp, by voice
Zero onboarding for the elder: WhatsApp is already on the phone. The rest of the family can message the home too.
-
Speak
She sends a WhatsApp voice note in her own language — the app she already uses every day.
-
Understand
Indian-language speech recognition and a language model extract intent, mood and context — including mixed Hindi–English.
-
Act
The agent plans and controls lights, fan, AC, plugs and door sensors directly, whatever the brand.
-
Reply
Confirmation comes back as a voice note in the same language — no reading, no menus.
SmartHomeAI · Home
Replies in Marathi
3:02 pm
0:03
“खूप गरम होत आहे”
“It is very hot.”
Checked before acting
- 34 °C
- 3 pm
- Her past preference
- Fan set to high
- Curtains closed
0:05
“खोलीत ३४ अंश तापमान आहे, म्हणून पंखा जोरात लावला आणि पडदे बंद केले.”
“It is 34 degrees in the room, so I have put the fan on high and closed the curtains.”
Fully autonomous, not just voice-controlled
The agent also acts without being asked, so routine comfort and safety never depend on someone remembering to ask.
- Lights on at dusk
- Light faded and fan off once everyone is asleep
- Geyser on before the usual bath time
- WhatsApp alert to the family if there is no movement by morning
Care alerts
Medicines, hospital visits and prayer times
The agent that runs the lights also keeps the day’s care schedule. Reminders go out as WhatsApp voice notes in the elder’s language, the lights can follow the five daily prayers, and if a reminder goes unanswered, the family hears about it. Watch one simulated day in two homes.
Starting the day…
SmartHomeAI · Home
WhatsApp
Today’s prayer times in your city
Calculated in your browser from the sun’s position.
Medicine reminders
- The family sets the schedule once: which tablets, and when.
- At each time, a voice note in the elder’s language, and the medicine-shelf light pulses.
- No reply after 20 minutes: a second reminder. Still nothing after 10 more: a WhatsApp alert to the family.
- The agent never changes a medicine or a dose. Only the family can edit the schedule.
Hospital visits
- The family adds appointments, or forwards the hospital’s message to the home on WhatsApp.
- Reminders the evening before and on the day: when to leave and what to take.
- The family hears the plan, and again when the elder leaves home.
- After the visit, a new prescription goes to the family to confirm before any reminder changes.
Prayer-time lights
- The five daily prayer times are calculated each day from the sun’s position for the home’s city, with no internet lookup.
- Five minutes before: a soft warm glow in the prayer room.
- At prayer time: the prayer-room light turns warm and bright, other lights dim, and a gentle reminder goes out. After 20 minutes, everything returns to how it was.
- A low hallway light at Fajr; lights stay off if nobody is home. Only for families who switch it on.
Sleep-aware rooms
Fell asleep with the lights on? The home notices.
When someone dozes off with the light, fan or AC running, the agent first checks whether anyone else in the room is still awake. If nobody is, it fades the light, turns off the fan, puts the AC into sleep mode and switches off a geyser left on, each the way the family chose. Change a setting and watch the night replay.
Starting the night…
Family settings · bedroom
Change one, and the night replays
Act after
Light
Fan
AC
Geyser
Saved so far tonight
0.00 kWh
Energy is an estimate against leaving everything as it was until 6:30 am: ceiling fan 75 W (40 W on low), LED light 12 W, 1.5-ton inverter AC about 1 kW at 22 °C with about 6% less power for every degree warmer (BEE), and a 15-litre geyser that loses about 0.8 kWh a day keeping water hot, half of which it spends again reheating in the morning.
Notices stillness
A small presence sensor, not a camera, sees that nobody has moved for a while and that breathing is slow and steady.
Checks the whole room
If anyone else in the room is still awake, nothing changes. The agent waits and checks again once they are asleep too.
Acts gently, the family’s way
The light fades slowly rather than snapping off; fan, AC and geyser follow the family’s settings. Anyone can override by voice note.
Daily electricity report
Every morning: what the home used, and what it cost
A short WhatsApp report arrives in the family’s own language: units used, money spent and the device that used the most. It comes with a plan for the day that needs no purchase and, from how this family actually uses each device, options for which devices are worth changing. Pick a family, a city and a language.
Preparing yesterday’s report…
Yesterday
Family of four · Bengaluru
Used yesterday
0 units
0 spent
Last 7 days
Hour by hour
By device
See the numbers as a table
SmartHomeAI · Home
Every morning at 8, in the family’s language
Today’s plan
No purchase needed. The agent does these through the day; switch off any the family doesn’t want.
Devices that suit this family
Worked out from how this family uses each device, with options. By type and star rating, never by brand.
Sample families with illustrative numbers. Rates: Bengaluru ₹5.80 a unit (domestic energy charge; fixed charges and taxes extra); Abu Dhabi 29.5 fils a kWh (expatriate rate above the first band). AC power falls about 6% for every degree warmer (BEE); a ceiling fan lets the AC sit about 2 °C warmer with the same comfort, and clogged filters can make an AC use 5–15% more, of which we count 5% (US Department of Energy). Ceiling fan 75 W, BLDC fan about 30 W; a 15-litre geyser kept hot all day loses about 0.8 units at 3 stars and under 0.57 at 5 stars (BEE). Switching a heater off is counted at half, because the tank is reheated later, and switching the AC off in an empty room at 60%, because the room is cooled again.
Measures each device
Smart plugs and connected devices report what each one draws and for how long. The agent turns that into units and money, hour by hour.
Reports in the family’s language
Every morning on WhatsApp, as a short message and a voice note, in whichever language the family uses: Hindi, Kannada, Arabic or English.
Plans first, then suggests
A plan for the day that needs no purchase, then device options with the saving for each. When keeping a device is the better choice, the report says so.
System architecture
Five layers, one brand-agnostic brain
Any compatible smart device, whatever the brand, plugs in through the perception and action layers without changing the brain.
-
Perception
WhatsApp voice notes, sensors, event streams
-
State & memory
Short-term context and long-term preferences
-
Reasoning
Goal-directed planning and tool selection
-
Action
Device control and user notifications
-
Reflection
Outcome checks and self-critique
Brand-agnostic by design
The reasoning layer never depends on a brand’s app. A new device only touches the perception and action layers.
Safety never waits on the cloud
A rule-based fallback runs locally, so safety-critical actions never wait on a language-model round trip.
Where things stand
Work completed so far
SmartHomeAI is at proof-of-concept stage, with a real device under the agent’s control. The next step is a whole-home agent in pilot homes.
Done — Phase 0, proof of concept
- Device cloud project set up for SmartHomeAI, hosted in India
- Physical test device: an RGBW smart bulb, paired and controlled by the agent
- Device-control software built and tested on the real bulb
- Agent codebase structured: device control, chat interface, language-model planner
- Reference architecture and teaching material on agentic devices authored
Next 6 months
- WhatsApp voice note → device action demo (Hindi, Marathi)
- Multi-device expansion: plug, fan controller, sensor kit
- Kannada and English; autonomous routines (dusk, empty room)
- First 5 pilot homes in Bengaluru
Why this is different — and why India
The intelligence and access layer, not more hardware
Global voice assistants are optimised for English-speaking markets and closed ecosystems. SmartHomeAI starts from the other end: Indian homes, Indian languages and the devices people already own.
Brand-agnostic by design
Works with the affordable smart devices Indian families already own. No new hardware to buy; the brain works with what people own.
Elders first, by voice, in their language
WhatsApp voice notes in Hindi, English, Kannada and Marathi — nothing to install, nothing to read. Built for the parent living at home, not the engineer who set it up.
Offline-first and resilient
Core routines run on a low-cost local hub; the cloud is an upgrade, not a dependency. Designed for power cuts and patchy networks.
Autonomous, explainable and safe
The agent looks after the home on its own and says why it acted, in a voice note. A family member can override anything; safety-critical devices use verified rules.
24-month roadmap
From one bulb to an open platform
Each phase ends with a demonstrable milestone: a working demo, a pilot report, a publication or an SDK release.
Phase 0 complete: a working proof of concept on a real device
-
Phase 1Months 1–6
Whole-room agent
- WhatsApp voice interface: Hindi, English, Kannada, Marathi
- Multi-device control (light, plug, fan, sensors)
- First autonomous routines: sleep-aware switch-off, opt-in prayer-time lighting
- Local hub prototype
-
Phase 2Months 7–12
Whole-home agent
- Memory and preference learning
- Care and safety alerts: medicines, hospital visits, no movement, door left open
- Daily electricity report and plan, in the family’s language
- Guardrails and family override
- Pilot: 5–10 elder homes, Bengaluru
-
Phase 3Months 13–18
Multi-agent home
- Lighting, HVAC, security and energy agents
- Orchestrator with conflict resolution
- Explainability dashboard
- Pilot: 50 homes across 2 cities
-
Phase 4Months 19–24
Open platform
- Developer SDK and device onboarding
- Research publications and open datasets
- Course module and trained cohort
- Partner integrations, commercial pilot
Commitments
Deliverables and impact by month 24
60+pilot homes running the agent
4languages via WhatsApp voice
3+peer-reviewed publications
1open SDK and device-onboarding kit
Energy
Tariff-aware planning and a daily electricity report in the family’s language target 10–15% lower household electricity use in pilots, measured against baseline months.
A target to be validated in the Phase 2 pilots, not a claim.
Inclusion
Elders living at home run the house by speaking naturally on WhatsApp in their own language, with no app to operate.
Education
A full course module and open teaching artifacts on agentic devices, grounded in a real deployed system.
Who is building this
A founding team across India, Germany and the UAE
Mathematics and production machine learning, applied research in Germany, and enterprise technology in the UAE.
Dr Jaya R.
Co-founder, Research & AI
Research Assistant, Hochschule Anhalt, Germany
Nabeel Bahussein
Co-founder, Technology & UAE
Head of Information Technology Department, Dubai Insurance Co.
Aman Deshpande
AI & ML Engineer
MS in Computer Science, Georgia Tech
Software engineer building production AI systems — agents, retrieval and the backend beneath them.
The team we are building
- Backend and IoT engineeringDevice integration, local hub, onboarding
- Applied machine learningPlanner, memory, evaluation loop
- Speech and language researchIndian-language intent understanding
- Pilots and communityHome deployments, user studies, feedback
Interested in joining or advising? Get in touch.
Coming to the UAE
SmartHomeAI in Arabic and English, from Abu Dhabi
An Abu Dhabi pilot from 2027 — for homes in the UAE, and for UAE families looking after their parents’ homes in India.
Work with us
Help bring SmartHomeAI into Indian homes
SmartHomeAI is preparing its first pilot homes in Bengaluru. We would like to hear from:
Families in Bengaluru
With a parent at home who would rather speak to the house than learn an app. Join the first pilot homes.
Register interestDevice makers and installers
Smart-device brands and installers who want an intelligence layer above their devices.
Discuss a partnershipResearchers and funders
Collaborate on agentic devices, Indian-language speech and open datasets.
Start a conversationDr. Milan Amrutkumar Joshi