​AI Token Burner Meetup is an invite-only, in-person gathering for AI builders in India who are actively experimenting with LLMs, AI agents, dev tooling, on-device AI, and voice AI.

​It is a room for engineers, researchers, developers, founders, and tinkerers who are deep in the weeds of building real AI systems: testing model behaviour, designing agent workflows, working with RAG pipelines, improving evaluation loops, optimising inference, exploring prompt and context engineering, and turning prototypes into products that can actually ship.

​The meet-up is designed around practical, builder-led conversations. Expect sharp technical discussions, live demos, and honest learnings from people working across the AI stack, from local-first inference and mobile deployment to agentic workflows, developer tools, and production-grade AI applications.

​For this edition, we’re bringing together a strong speaker lineup with hands-on experience building and deploying AI systems in the real world.

​

​Speakers:

​Shivay Lamba, Lead ML Engineer at Qualcomm :

​Shivay will be speaking on “Running Local-First, On-Device LLMs on Android with QNN using LiteRT & ExecuTorch.” His session will explore how real AI models can run fully on-device, covering architecture, deployment pipelines, NPU performance tuning, and demos across modern AI PCs and mobile devices. The talk will focus on what it takes to build fast, private, low-latency AI experiences without depending entirely on the cloud.

​More speakers will be announced soon, each bringing a different lens on where AI development is headed next.

​If you are building with LLMs, agents, inference stacks, voice AI, or developer tooling, this meet-up is for you. Expect a room full of people who are not just talking about AI, but actively burning tokens, breaking things, testing ideas, and building what comes next.