Artificial Intelligence Building for All: Lessons from Kaustav Das Sharma and Gumlooop
Obstest artificial intelligence everywhere. Chatbots and virtual assistants respond to customers, and large language models (LLMS) help in making blog publications, viral videos and even student education. But just because it is widespread, it does not mean that it is easy to use.
Many companies struggle to adopt artificial intelligence because of their technical complexity, high costs, a lack of talents needed to work with them, and the problems of expansion, among other things.
to
Kaustav Das Sharma and its impact on the development of artificial intelligence
The career path of Sharma was formed by a childhood that has eliminated action all over the world. He was born in Calcutta, lived in cities like Milan, Dubai, Stavenger, and London before heading to Montreal to the university.
He taught him to move regularly quickly and communicate with people from all aspects of life, but he was aware of the tremendous contradiction in the quality of life between the places where he lived and where he was born.
See these disparities at an early age inspired Kaustav to make a change. A rupture between the study of economics and engineering, ultimately chose engineering, not only to build things but to create practical solutions that make life easier for others – especially those who do not get resources.
This belief that talent is universal, but the opportunity is not, still directs his work. From his time in Amazon Web Services (AWS) to his role as the first engineer in Gumlooop, Kaustav has constantly designed tools that facilitate the adoption of artificial intelligence, giving more people the opportunity to create and innovate.
In AWS, Kaustav faced a major challenge through Aurora Hackathon: Simplify how engineers write to database inquiries. Aurora, a tool that helps institutions to manage data such as sales tracking or storage control, requires engineers to use SQL (organized inquiry language), which may be difficult for those who are not database experts.
Therefore, KaUSTAV and his teammate built the AI tool that allows users to describe their data needs in a clear language, and converted these descriptions to SQL queries automatically. They worked hard to make the tool reliable, clean the data you used and switch its accuracy with human comments and tests.
Ultimately, the tool was easy to use, quickly writing the query, and made the database tasks easier – even for KaUSTAV, which had a limited SQL experience at the time. Their work won the infiltrators and became a vital resource for the team.
After this success, KaUSTAV and his teammate Rahul, co -founder of Gumlooop, have focused on education, as they faced a different kind of challenge.
Traditional lectures, where students listen and play, has not changed much for decades. “After the ruling and in the era of a mobile phone, children are excessive, and they are struggling to focus, and traditional lectures are not effective for learning.”
To address this, they built
Then KaUSTAV and Rahul continued to face a problem closer to the home: how difficult people in the field of mental health care, even in countries that have a lot of resources. They found that treatment often feels far away. Expensive, confusing, or frustrated.
For help, they built Therapai, an Amnesty International tool designed to facilitate the process by allowing users to set preferences for the type of processor they want, whether someone listens to or a more focused on solutions. It also contains guarantees, such as clear approval to use data and safety standards to link users to real health care providers.
Although they were unable to completely developed Therapai because of time restrictions – Rahul was busy with YC and Kaustav was in AWS – the project taught them a lot. The demand for artificial intelligence tools has shown easy to use and design, taking into account the needs of the user.
KaUSTAV took these ideas to work in Gumlooop, where he continues to make technology practical, available and useful for ordinary users.
Kaustav’s five lessons for the AI process that can be accessed
In Gumloop, Kaustav Gummie, a virtual assistant over 24/7 answers questions and solves workflow problems. Users can describe what they need in a clear language, such as “summarizing this document and sending it by email to my team”, and Gummie takes over the rest. This approach provides time and makes project management easier for people who have no technical background.
Gummie has already made Gumloop easier for users, shortening spare time and reducing manual customer support. “Imagine your imagination to exist,” says Kaustav. “This is what Jumi aims to achieve – turning ideas into actions with artificial intelligence.”
The construction of obstetric intelligence tools such as Gummie, Studyguru and Therapai tended Kaustav some important lessons on how to simplify technology for ordinary users.
Here are some of the main fast food:
Lesson 1: Quick charging, repetition, and focus on users first
When creating a new product for Amnesty International, Kaustav believes that the key is to launch a product of no less than a product (MVP) – a basic version of the product that has enough features to be used by the first customers – quickly. The goal is to get everything perfect immediately but to get something in the hands of users, collect comments, and make improvements based on the real world use.
Lesson 2: Adoption of the main principles from the beginning
KaUSTAV proposes two important principles for any new Amnesty International project: separating different services in your code so that it is easier to adapt later, and avoid making unimaginable decisions. By quickly moving with the decisions that can be reversed, you keep your team flexible and you can avoid stumbling in the long run.
Lesson 3: Look for your community and display your work
Kaustav’s profession started when he and Rahul began to communicate with the broader artificial intelligence community. The attendance of the conferences and the participation of their work not only gave them valuable reactions, but also presented them to people who will play a major role in their future, such as Max, who is now the CEO of Gumlooop. “The ocean yourself with a society similar in thinking can make a difference,” says Kaustav.
Lesson 4: A great idea of pushing the global influence
KausTav believes it is easy to stumble in focusing on small specialized solutions, but real innovation occurs when you think more. The goal is to create AI products that can change industries and make a permanent impact on a wide range of users.
Lesson 5: The role of feedback in optical design
Kaustav’s previous projects, such as Therapai and Studyguru, put the user’s needs in the forefront and the center. While working on these tools, he allowed him to listen and implement user notes to build with a practical goal, which puts the way for more harsh projects such as artificial intelligence addiction with Gumlooop.
Directing a portal of artificial intelligence
While a lot of people play with obstetric artificial intelligence in their daily lives, taking it on a broader scale still can only feel experts who experts can withdraw.
Fortunately, the vast Sharma Engineering experience and unique education helps make this technology easier for everyone, regardless of their background or experience.
Kaustav Das Sharma imagines that Gumlooop has become a platform that transforms how people interact with technology, and the complex workflow has turned into simple and intuitive processes. It aspires to build systems that enable workers in all industries to benefit from artificial intelligence without the need for artistic experience.
To learn more about the latest Gummie and Kaustav projects, check
