name

Sohan Maheshwar

Sohan is a Developer Advocate based out of Amsterdam, Netherlands. He is deeply passionate about emerging technologies and how it is shaping the world around us. He frequently works with developers in startups, ISVs, and enterprises on their cloud strategy. He previously worked as an Alexa Evangelist at Amazon, and has been in the developer relations space since 2013.

Outside of work, Sohan loves playing frisbee, listening to rock music and reading comic books.

Talks

  • Decoding the hype behind WebAssembly: Looking under the hood

    WebAssembly is garnering a lot of excitement because of it's unique combination of portability, security, and performance. We're seeing it used widely across browser, server-side, and IoT use cases. But what actually is WebAssembly? This talk will crack open the hood of WebAssembly and help demystify what is going on inside. The core of the talk will focus on understanding what a WebAssembly module is and how it is executed. We'll explore the text and binary representations of a module; the data layout within a module; and understand the basics of how its bytecode instructions operate a stack machine. Then with an understanding of what a WebAssembly module is we'll take a look at how languages are compiling into WebAssembly and the types of runtimes that are actually executing WebAssembly. The talk will conclude with some of the usecase adoptions of this new technology.

  • Securing RAG Pipelines with Fine Grained Authorization

    It’s time for Day 2 Ops in the world of AI. Building enterprise-ready AI poses challenges around data security, scalability, and integration, especially in compliance-regulated industries. Firms are increasing efforts to mitigate risks associated with LLMs regarding sensitive data exfiltration of PII and sensitive company data. The primary mitigation strategy is to build guardrails around Retrieval-Augmented Generation (RAG) to safeguard data while also optimizing query response efficiency. This session will cover how modern permissions systems can safeguard sensitive data in RAG pipelines. We'll start with why Authorization is critical for RAG pipelines to protect sensitive data from potential vulnerabilities and also the various techniques for permissions-aware data retrieval including prefiltering, and post-filtering vector databases. The talk will also include a practical demo implementing fine-grained authorization for RAG using Pinecone, Langchain, OpenAI, and SpiceDB.