Kaushik Bokka
ML Consultant / Advisor
India · 1 connectionNot on Unlinked · Imported profile. Is this you? Claim it
Is this about you? Request removal
About
I build systems where models meet meaning. Over the past few years, I’ve worked across open source, devtools, consumer, and multimodal AI — from designing frameworks that help developers integrate Visual Language Models into production, to building narrative-aware agents that understand long videos and movies. Currently, I advise Rumi Labs (an a16z-backed startup) on long-video understanding and eval systems, where I’ve built the RumiEvals SDK — automating multimodal benchmarking, retrieval, and reasoning across temporal, narrative, and emotional dimensions as well as the RAG system with Dual Context Retrieval Strategy. Previously at VLM Run, I helped design the MCP Server, which enables visual processing for LLM agents, and built scalable video pipelines using open-source VLMs. Before that, I co-founded Sequels AI, an Agentic AI Television OS, which generated contextual, scene-aware content for live media — later licensed to an a16z-backed company. Earlier, I contributed to the open-source ML ecosystem through PyTorch Lightning, Guardrails AI, and Uber Ludwig, working alongside teams at Meta, Google, and Salesforce to bring distributed and multimodal capabilities into mainstream M
Experience
Education
Technical Strategy Program
Bachelor's degree, Electronics and Communications Engineering