WhatsApp Image 2019-02-10 at 5.40.43 PM.

Rahi, a System to Address Sexual Harassment in Urban Colleges in India



UN Women





Madhushree Kulkarni


UN Women

Rahi is a very personal endeavour which began as a zine but has now become a full-fledged systems design project. We will be collaborating with a Jhatkaa.org, UN Women and a few more subject matter experts and organisations to make this project a reality. 

Note: The Research documented here has been done before coming up with a definite problem space. We had no idea that something that started out as a simple zine would turn into something so big.  

Keywords - Research, Systems Mapping, Sexual Harassment, Systems-Oriented Design, Systems Design


When one of our core team members faced sexual harassment in college and raised an alarm, she found that there was no one to back her up. The college did have an ICC but it was barely functional. Moreover, the attitude of the authorities seemed to be to brush these complaints aside instead of trying to enforce safety norms. In the light of the recent outings of college students on the Instagram page, "Humans of Patriarchy", we realised it's a problem that plagues most of the colleges in India. So we decided to do something.


Rahi is a system with three major aims:

  1. To provide survivors with a tool to identify / report predatory or unwanted behaviour in their campuses.

  2. To create a network of legal professionals and NGOs that help enforce safety norms for these campuses.

  3. To create a network of mental health professionals which guide the survivors and provide rehabilitation to minor perpetrators who haven’t committed major offences

  4. To design workshops for colleges to educate


A. Part 1 (2018) - 

Initially, we made a zine that was aimed at female students. The zine spoke about different protective laws and illustrated them through the medium of comics based on issues that women tended to face on a daily basis. We also added a FAQ section which addresses the questions which falls under the grey area. 

matchstick zine-01.png
matchstick zine-02.png
faqsss zine-01.png
faqsss zine-02.png

Although the aim while creating this was just education, we realised we wanted to do more. Especially after facing harassment ourselves, we realised that publishing a zine won't make a lot of difference. Hence, we decided to dig deeper.

B. Part 2 (2019) - 

while researching we started noticing a pattern. We realised that we couldn't assess if something was right or wrong because we couldn't label the behaviour or incident. We wanted to confirm this by conducting primary research. We started by sending out google forms that collected incidents of harassment/ abuse/ gaslighting/ discrimination/ bullying. After which we had discussions with almost every single girl in the hostel and students from FLAME University, Pune. That along with secondary research helped us understand the problem we were dealing with properly and helped us move forward. We collaborated with subject matter experts, NGOs, activists, lawyers and engineers. We conducted a survey to find out more. After realising most of the people faced this issue, we decided to create an app which lets you identify behaviour and report it.

WhatsApp Image 2019-02-10 at 5.40.43 PM.
WhatsApp Image 2019-02-03 at 3.57.18 PM.
WhatsApp Image 2019-02-03 at 3.57.16 PM.
Screen Shot 2020-05-13 at 5.22.43 PM.png
Screen Shot 2020-05-13 at 5.23.21 PM.png
Screen Shot 2020-05-13 at 5.22.56 PM.png
Screen Shot 2020-05-13 at 5.28.09 PM.png

Results -


1. We learned that 82% of all women find it difficult to label incidents and behaviours. One of the reasons being that we have been normalised           believe something is normal or acceptable even when it's not.  

2. Only 3% of the girls tried reporting the violence to college authorities.

3. Out of the ones who tried reporting, they didn't know how to go about it and were confused. 

3. 30% of them got professional help. 

4. 90% of the girls did not know what the ICC is or what it entails. 

5. Nobody was aware of any kind of laws at college campuses. 

We mapped out the current system and identified the obvious barriers. We will be doing our research and mapping out these barriers is detail. 

1. The normalisation of rape culture

2. The inability to identify predatory behaviour due to lack of education

3. Influence of social groups, peer pressure and gaslighting

4. Complicity of authorities

5. Fear of retaliation

6. Social pressure - fear of being judged, shunned or punished.

7. Lack of awareness about mental health and support systems

It is a pity that most of us in our country are never even taught to differentiate between good and bad touch, as part of our childhood grooming, let alone having an open discussion about matters related to sex or harassment. What it leads to, in turn, is a confused reception by the victim, of any form of sexually inappropriate behaviour. In fact, there are people who do not even recognise sexual harassment initially, even after being subjected to it. The more we researched we realised the complexity of this problem. But, we didn't want to stop here. We continued our research and reaching out to experts religiously taking notes and having meetings every month. 

C. Part 3 (2020) - 

Rahi/ राही:

Traveller; Companion.


This complex and hostile environment against women is a product of deeply entrenched patriarchy; an outdated and rigid educational system; caste system; longstanding religious and cultural tradition; weak institutional support; and social stigma against, and collective denial of the issue of sexual assault.

Large and sustained change cannot be achieved with short-sighted, one-off, or reactionary changes. Widespread behavioural change can only be achieved if the solution landscape targets education, policy, law, technology, and infrastructure.

Eventually, we decided to take the systems thinking approach and we decided to create a whole system surrounding the app. We decided to reach out to subject matter experts and organisations for possible collaborations to undertake this huge project. And surprisingly, we heard back from Jhatkaa.org and put together a plan to go ahead with the Project. I also participated in a UN Women Challenge in August and spoke about the Project to Melissa Alvarado, who is a Program Manager for UN Women. She and her team will also be helping out with the Project.  We realised that for an app or a system that would be used in an Indian context, we would need a name familiar to people. The journey to ending sexual harassment and rape culture in campuses is long and arduous. We hope to provide a trustworthy support system to the ones taking this journey, hence the name Rahi.

Asset 1.png


When we reached out to organisations and SME's, we created a deck explaining the problem and the solution space and the approach that needs to be taken. We spoke in detail about the app and other factors surrounding it. We created a few prototypes to visualise what the concept ot the app could look like-

Affinity mapping Teachers (1).jpg

The an app uses a bot to provide an online safe platform for students to document and report their sexual assault. Talking to the bot is like texting or using a messaging app, but with a machine instead of a friend. It’s an unbiased bot, available 24/7. This survivor-centred approach increases the chances that a survivor will report and that they will experience positive emotional and adjudicative outcomes. It also enables schools to detect repeat perpetrators. It offers three options:

1. A tool to help you decide and label whether a behaviour/ incident was abuse/ violence/gaslighting/ invasion of privacy/   


2. It simplifies the process of reporting harassment.

3. An educational tool, where you can read up about policies, laws, organizations and people working in this field. 

4. A platform that lets you connect with professionals

We learned from the best practices of apps already created with a similar concept - ai to report sexual harassment and are trying to develop one to suit the Indian context.

Why a bot?

Students are often afraid to report harassment. Because some students are unaware themselves and don't know how to label incidents, a bot is helpful because:

1. It'll never judge or assess you.

2. It's available anytime, anywhere, for as long as you need.

3. It won't share your conversation with anyone unless you explicitly tell it to do so.

Cognitive Interview

The Cognitive Interview was developed by scientists to help police reliably extract accounts of important life events. Decades of research show that this technique increases the amount of accurate information and decreases the amount of inaccurate information obtained in an interview. The Cognitive Interview is currently considered best practice for interviewing people about negative emotional memories. 

The Cognitive Interview works by first asking you to describe freely what happened, followed by specific questions called ‘probes’ that ask you about the things you mentioned.

Part of what makes this technique effective is the way in which the questions are asked. All questions are open-ended rather than leading the person to a certain response. It takes substantial knowledge to create a high-quality set of questions.


NLP is a branch of informatics, mathematical linguistics, machine learning, and artificial intelligence. NLP helps the bot analyze the human language and generate the text to interact with the user. The bot identifies important details mentioned in a student's account and then probes more deeply. This type of probing reflects the cognitive interview practice of asking open-ended questions based on details mentioned by the person being interviewed. Since humans tend to introduce bias, a bot combined with NLP may be able to conduct a cognitive interview more rigorously than a human.

Human Intervention

No matter how much of advanced technology is used, we believe tackling sensitive problems like these will need human intervention. We know the bot is after all a machine and prone to making errors. That is why, when you are trying to label an inside, the bot doesn't give you an answer but gives you the tools necessary to make a decision yourself. While reporting an incident or behaviour, the bot only asked you open-ended questions and gives you options on how to take action. 

The app also connects perpetuates and victims to counsellors and therapists. 


1. Forming a team of subject matter experts and organisations to work on this project.
2. Starting with the Research.
3. While the app would help people label behaviour and report the assault, we found ourselves asking these questions:


  • What were the chances that the reports would reach the right authorities? 

  • What if the authorities themselves were complicit?

  •  What languages are people comfortable with, especially when it comes to personal issues? 

  • Would they rather use audio than text? If yes, how will the audio be parsed? 

  • What person at an NGO will be responsible for dealing with the reports? How do they expect to receive information? 

  • Who will have access to the data? How would we maintain privacy?

  • How would we deal with anonymous reports and false reports?

  • What biases would creep in?


In our efforts to find solutions to these questions, we realised the enormity of the task that lay ahead of us. Although we have a few blueprints for solutions, we acknowledge that there are a lot of factors which we still have to delve into and tweak our solutions accordingly.

4. There is a possibility that all of this research might not be helpful or the solution space will need to be worked on after we undergo te secondary and primary research. 

Feel free to reach out for a collaboration or a conversation!

© Aishwarya Jare