Akshata Mohanty
Founder, Keis
Singapore
Actions
Akshata is a cloud architect with a focus on data applications and AI agent automations. She is a Google Developer Expert (GDE) in Artificial Intelligence and Machine Learning - Generative AI. She also founded Keis, a tech consultancy that helps enterprises build cloud-native data solutions. She currently works as the Lead Engineer at Tilla Technologies GmbH, and has previously held roles at NasAcademy, Rakuten, and Flipkart.
Akshata graduated from IIT Kharagpur in 2015 with a Bachelor's degree in Architecture. She began her career in research at the National University of Singapore, before transitioning into the tech industry, where she combines technical depth with a strong product and systems-thinking approach.
Area of Expertise
Topics
Architecting AI Agents: Supercharge your workflows with Vertex AI
Curious if AI could really replace you? In this talk, we dive into the fast-evolving world of AI agents and their powerful capabilities. Starting with a comprehensive introduction, we'll explore the core concepts involved in designing the system powering an AI agent. We then discover the different architectures leveraging Vertex AI, and evaluate the challenges and solutions for each. Finally, we'll wrap up with some real-world examples and a live demo, showcasing the power and versatility of AI agents in action.
Good (n)gram Hunting: Building Serverless Search at Zero* Cost
This talk walks through the process of transforming an unstructured database of over 30 million entities into a text-searchable resource. Instead of using an existing solution, we built our search in-house using ngrams, serverless computing and object storage. We will deep-dive concepts of indexing, (n)grams and tf-idf. And explore the insights from the successes, challenges and limitations of this project.
Good (n)gram hunting: How we built search in-house at (almost) zero cost
This talk walks through the process of transforming an unstructured database of over 30 million entities into a text-searchable resource. Instead of using an existing solution, we opted to build our search in-house using ngrams, serverless computing and object storage. We will deep-dive concepts of indexing, (n)grams and tf-idf. And explore the insights from the successes, challenges and limitations of this technical project.
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