Currently working on new proptech AI ideas. Senior Software Engineer previously at Opendoor, I was dedicated to revolutionizing the title and escrow industry through innovative technology and automation. With over 7 years of experience in developing and scaling microservices using AWS, Node.js, and serverless frameworks, I have led teams in integrating tech stacks, overseeing legacy integrations, and planning migrations. Previously, I served as a team lead at JetClosing, focusing on building microservices for various functionalities. My background also includes work with Microsoft's artificial intelligence & research division, where I automated security tools and developed a web app for CI workflow using Azure DevOps Services.

AI agents, AI Eval, RAG, proptech, functional programming, microservices, event driven architecture

AI agents to stream real estate grunt work that is tedious and highly error prone.

Ravi Gupta is a Technical Lead at AMD with over 10 years of experience in software engineering and hardware design. He specializes in pre-silicon simulations and developing software stacks for custom hardware SOC designs. Ravi holds a Master’s degree in Computer Engineering from Purdue University and a Bachelor’s degree in Electrical and Electronics Engineering from Manipal Institute of Technology. He has strong coding skills in C++ and Python, with expertise in AI model acceleration, deep learning, and cloud-native data analytics. Passionate about building strategic alliances and personal development.

Distributed inference, studying collectives, PyTorch internal, MoE placements and enhancements

vLLm and SGLang on AMD GPUs and system level debugs for MoE models. In the past , I have worked with Large National Labs where I have scaled computations for Models up to 171 nodes(1026 GPUS)

Currently exploring AI agents, MCP, local LLMs. 5+ years of experience working on high-availability, low-latency distributed systems at AWS, both building new customer functionality and working with legacy systems. Currently working on expanding functionality of HumanInLoop systems at AWS

- AI agents and furthering functionality with MCP - Local LLM - HumanInLoop workflows, particularly with how context and prompts can make humans more effective.

- Building AI agents that coordinate among multiple peers via WebRTC, pulling context from calendar MCP, and using open source TTS models - Using MCP tools to automate regular household management