JalTaap: Making Heat and Water Safety Accessible to Every Indian Worker
Built by Team DevCore at the WeMakeDevs × AWS Bharat Builds Tour hackathon at DTU, JalTaap transforms weather data into personalised, multilingual safety guidance for outdoor workers while connecting local problems to the authorities who can solve them.
JalTaap: Making Heat and Water Safety Accessible to Every Indian Worker
What if a weather warning could speak your language, understand your location, and tell you what to do next?
For millions of outdoor workers in India, extreme weather is more than an inconvenience. It directly affects their health, safety, and livelihood. Construction workers, delivery riders, street vendors, sanitation workers, and others often have to continue working through intense heat, heavy rain, and difficult conditions. Yet the information designed to protect them is not always personal, accessible, or actionable.
This challenge inspired JalTaap, a heat, rain, and water safety assistant developed by Team DevCore during Environmental Hacks, part of the WeMakeDevs × AWS Bharat Builds Tour, held at Delhi Technological University (DTU).
Our goal was simple: make environmental safety information understandable and useful for the people who need it most.
The Problem: A Warning Is Only Useful If People Can Act on It
Traditional weather alerts often communicate general conditions across an entire city. But a city-wide warning cannot tell every worker when conditions at their exact location may become dangerous, when they should take a break, or where they can find drinking water and relief from the heat.
Language and accessibility create another barrier. Information presented in English-heavy apps or text-based alerts may not reach people who are more comfortable speaking their regional language or who have difficulty reading.
There is also a gap between reporting a problem and getting it resolved. Flooded roads, water leaks, and missed water-tanker deliveries can affect entire communities, yet reports may not reach the department responsible for taking action.
We saw an opportunity to connect these problems through one accessible platform.
Introducing JalTaap
JalTaap is designed to turn environmental data into practical, personalised guidance. Rather than simply telling users that the weather is hot or rain is expected, it helps them understand what those conditions mean for their immediate safety.
One of our central design principles is that users should be able to interact in a language they understand.
JalTaap supports Hindi, Hinglish, and English across its conversational experience. Users can communicate through WhatsApp, Telegram, or the website. They can type a question, share their location, or send a voice note. Responses can also be read aloud, making the experience more accessible to people who prefer listening over reading.
This approach puts accessibility at the centre of the solution rather than treating it as an optional feature.
How JalTaap Works
JalTaap brings together four key stages:
1. Ask
A user sends a message, shares a location pin, or records a voice note. The system is designed to make accessing safety information as straightforward as starting a conversation.
2. Calculate
JalTaap uses forecast information for the user's location to assess hourly risks associated with heat and humidity, hot nights, prolonged heat, rainfall, air pollution, and strong sunlight.
An important distinction is that the risk calculation is handled by a tested, rule-based calculator. AI helps communicate the results, but it does not independently invent the risk level.
3. Answer
The user receives clear, location-specific guidance, including safer working hours, when to take a break, hydration advice, and nearby cooling centres or other relevant facilities.
The interface prioritises simple language, visual indicators, and voice responses. For emergency situations involving signs of heatstroke, it prominently directs users to call India's emergency number, 112.
4. Act
JalTaap also addresses environmental problems beyond personal safety. Users can report flooding, water leaks, and tanker shortages. Reports can be checked, mapped, and routed to the relevant authority based on the location and type of issue.
This creates a connection between people experiencing a problem and the organisations responsible for addressing it.
Why Humidity Changes the Picture
One of the most important insights behind JalTaap is that temperature alone does not tell the whole story.
High humidity reduces the body's ability to cool itself through sweating. As a result, conditions can become dangerous even when the temperature has not crossed a conventional heatwave threshold.
In our backtesting of modelled weather data from April to July 2026, we examined this gap across 63 Indian cities.
Our analysis identified 414 city-days with dangerous humid-heat conditions that a simple 40°C threshold would have missed.
The findings were particularly striking in some cities. In Bhubaneswar, our analysis identified 50 dangerous days, while the 40°C rule caught only three. In Kolkata, it identified 42 dangerous days, none of which met that temperature threshold.
These results highlight why a more comprehensive approach to heat-risk assessment matters. They are based on modelled weather data from Open-Meteo, rather than direct weather-station measurements, and should be understood within that limitation.
For us, the lesson was clear: environmental warnings should consider the conditions people actually experience, not temperature alone.
Built with AWS: Technology with a Purpose
JalTaap uses AWS services and open-source tools to support its conversational interface, data processing, reporting workflow, and scalable architecture.
The project incorporates Amazon Bedrock for AI-powered conversations and image checks, AWS Lambda for serverless processing, and Amazon API Gateway for API access. Amazon DynamoDB and Amazon S3 support data and file storage, while Amazon SQS and Amazon EventBridge Scheduler support asynchronous processing and scheduled tasks.
Accessibility is also part of the technical design. Amazon Transcribe supports voice-input workflows, while Amazon Polly supports spoken responses. AWS Amplify and Amazon Cognito support the web experience and authentication, and AWS CloudWatch supports monitoring.
We also incorporated tools such as Strands Agents, AWS Serverless Application Model (AWS SAM), AWS Lambda Powertools, and Cedar for agent workflows, infrastructure-as-code, operational practices, and access-control policies.
The purpose of this technology stack is not simply to add AI to a project. It is to use cloud infrastructure and automation to make environmental information more accessible, useful, and scalable.
Designed for People, Not Just Screens
A safety platform must be usable under real-world conditions.
That is why JalTaap uses a picture-first interface with prominent visual indicators, short messages, and a Listen option. Users can access the conversational experience through WhatsApp without needing to install a separate application or create an account for that interaction.
For supervisors, JalTaap includes crew-level heat planning to help organise work hours, rest breaks, and water requirements.
For authorities, it provides a reporting workflow with mapped issues, assigned departmental queues, and action lists. This is intended to help ensure that reports do not simply disappear into a general complaint channel.
Together, these features extend JalTaap beyond a weather assistant into a platform connecting individual safety with community-level environmental action.
Building, Testing, and Learning
Developing JalTaap required us to think beyond the initial idea. We had to consider how users would communicate with the system, how environmental risks would be calculated, how reports would be routed, and how the platform could remain transparent about its data and limitations.
Our project documentation records 196 automated tests passing, alongside 34 test conversations used to check the AI's responses. The risk calculations were also tested against published reference values.
We recognise that testing a system is not the same as proving its effectiveness in real-world deployment. Forecasts can be imperfect, local flood information needs verification, and JalTaap is intended to complement—not replace—official warnings and emergency guidance.
Being transparent about these limitations is an important part of building a responsible solution.
What Comes Next?
JalTaap is designed with expansion in mind. Future priorities include broader official weather-warning integration, additional Indian languages such as Bengali, Tamil, Marathi, and Telugu, SMS access for people without WhatsApp, and partnerships with employers and city control rooms.
These steps could help extend the platform's reach and improve the quality of local environmental information.
Our long-term vision is to make it easier for people to understand environmental risks in their own language, wherever they live and work.
More Than a Weather Alert
Building JalTaap with Team DevCore at the WeMakeDevs × AWS Bharat Builds Tour hackathon at DTU reinforced an important idea for us: technology has the greatest value when it addresses a real human need.
A weather forecast provides information. JalTaap aims to turn that information into a decision a worker can understand and act upon. It combines personalised risk assessment, multilingual communication, voice accessibility, and a route for community problems to reach the appropriate authorities.
Because safety should not depend on the language you speak, the phone you use, or your ability to read a warning.
That is the idea behind JalTaap—and the impact we hope to build towards.
Built by Team DevCore for Environmental Hacks 2026, Heat & Water track, as part of the WeMakeDevs × AWS Bharat Builds Tour.
Data note: The heat-risk figures cited above are based on the team's backtesting of modelled Open-Meteo weather data for April–July 2026. JalTaap provides supplementary guidance and does not replace official meteorological alerts or emergency services.
Enjoyed reading this content? Let the author know!
Your likes, comments, shares, and saves help creators reach more builders.
Loading recommendations
Loading article