Optimizing AI Pipelines: Parallelism, Caching & Streaming

​About Event

​Are you building AI applications but struggling with slow responses, inefficient workflows or rising infrastructure costs?

​Hosted by Pragya Singh, this hands-on workshop is designed for developers, AI enthusiasts, and builders who want to create faster and more scalable AI systems. Discover how techniques like parallelism, caching, and streaming can significantly improve AI pipeline performance through practical examples and collaborative exercises using Miro.

​What to expect:

• ​Hands-on workshop on optimizing AI pipelines using parallelism, caching, and streaming

• ​Interactive exercises using Miro to redesign AI workflows and identify performance bottlenecks

• ​Real-world examples of production AI systems, LLM applications, and agentic workflows

• ​Best practices for reducing latency, improving scalability, and enhancing user experience

• ​Networking with local developers, AI enthusiasts, and builders working on cutting-edge AI applications

• ​Open discussions and Q&A on AI architecture, optimization strategies, and engineering challenges

​Whether you're building AI agents, RAG applications, chatbots, or LLM-powered products, join Kolkata's developer community for an afternoon of learning, collaboration, and practical AI engineering.

​Organized by:

​Pragya Singh

​Event details:

​📍 IdeaPod Parkstreet Kolkata 📅 30th October 2026 ⏰ 1 PM - 3 PM