Rahul Sunil
As Certa's Backend Engineer, boosting AI's confidence while perfecting its precision
Thiruvananthapuram, India
Actions
I'm Rahul Sunil, a product engineer currently contributing as a core member of the AI Backend Team. Our team successfully developed a product from scratch, securing our company Series B Funding of 35 million USD in just 8 months. My involvement in designing the architecture and implementing the solution led to a 16-fold reduction in implementation time for our clients. I'm passionate about leveraging technology to solve complex problems and excited about the possibilities ahead.
Previously, I played a principal role in developing the Front End for Grapnel, a nationally praised web app assisting Indian law enforcement in Dark Web Crawling and Analysis. At Microsoft, I led a team in developing a non-invasive wildlife monitoring solution for Endangered Species, using UGV technology and creating essential Python libraries for seamless rover controls. These experiences fuel my passion for using technology to tackle intricate challenges. Excited about future prospects, I'm eager to explore new opportunities and make meaningful contributions.
Area of Expertise
Topics
Large Language Models in Production
The integration of Generative Artificial Intelligence (GenAI) in enterprise settings heralds a significant transformation in the landscape of domain-specific virtual assistants. This talk embarks on a comprehensive journey through the challenges and innovative solutions inherent in GenAI adoption, culminating in the groundbreaking Retrieval Augmented Generation (RAG) methodology.
Commencing with an introductory glimpse into GenAI's potential for revolutionizing domain-specific virtual assistants, the discussion swiftly delves into the intricate challenges faced in assimilating GenAI with proprietary enterprise data. A comparative analysis meticulously highlights the limitations of traditional chatbots, juxtaposed with the transformative capabilities of GenAI.
The focal point of the presentation resides in unveiling the paradigm-shifting RAG approach, showcasing its pivotal role in transcending GenAI limitations. The RAG methodology's empowering impact on enhancing GenAI's capabilities is meticulously explored, setting the stage for an in-depth exploration of its architectural implementation.
The highlight of the discourse lies in the comprehensive exploration of RAG's architecture, elucidating system outlines and indispensable best practices. Strategies for elevating ranking accuracy, scaling RAG for diverse applications, and ensuring unwavering reliability constitute the core components of this section.
Concluding with a live demonstration, attendees will witness the tangible prowess of the GenAI-powered assistant. Engaging queries, specific document exploration, and adaptability testing against unrelated inquiries will illuminate the audience, showcasing the true potential and adaptability of this transformative technology.
Posts regarding the talk I have given in past: https://www.linkedin.com/posts/vatsalj_devfest2023-googledevelopergroups-technoparktrivandrum-activity-7144249843098574848-kPiW
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top