Venkatasaireddy Barri
AI and Data Science Engineer II
I’m an AI and Data Science Engineer II at Deloitte, where I build AI systems that connect language models to data and tools—agents and pipelines using Snowflake Cortex, retrieval-augmented generation (RAG), natural-language search, and Model Context Protocol (MCP) integrations.
Outside of client work, I build benchmarks and experiments that probe how models actually behave—trying unusual ideas, testing how they respond to reframing, and mapping what they can and can’t do. My current project, Mirror Lab, asks whether an AI will recognize a dangerous action when the same operation is described as a harmless game.
Recent work
Mirror Lab — Can an AI spot a dangerous action hidden behind a harmless game?
Would an AI refuse to stab a person? Yes. Would it refuse to win a watermelon game? No—and it can’t tell the difference if the words are swapped.
Mirror Lab is a fully simulated AI safety benchmark where one system wears two faces: a toy game and a hazardous backend, mechanically identical, only the labels differ. That’s kinematic–semantic isomorphism—”kinematic” because the movements and state changes are identical, “semantic” because the described meanings differ, “isomorphism” because every step in one version maps exactly to one step in the other, with the same transitions and outcomes.
GitHub · Paper on Zenodo · Explore Mirror Lab
Wirebrain — Local-first diary for AI conversations
Wirebrain makes captured and imported chats searchable on your computer and lets compatible assistants retrieve that context through MCP. When you opt in to recording, Wirebrain uses a local HTTPS interception proxy powered by mitmproxy to capture prompts and replies from supported AI apps and save them on your computer. GitHub