Speakers
Our event will be In Person considering that conditions provided by Indian laws and the venue are satisfactory for running the event without any risks to the health of attendees, sponsors, speakers, and organizers.
Talks
Big ideas. Real experience. Meet the minds behind the talks.
- Latest trends in JS and AI
- Network with open-source creators
- Upgrade your skills
- Network with top developers
Showing 8 of 14 sessions
Harness the power of JavaScript Proxies.
Javascript Proxy is one of the lesser-known features introduced in ECMAScript 2015. In this session, we'll explore the benefits of using Proxies and discuss the types of problems they are best suited to solve. I'll share my own experience of using Proxies to create an assertion library for the Vest validation framework. I'll explain the challenges I faced with traditional methods and how Proxies allowed me to overcome them. Together, we'll build the assertion library step by step, so you'll have a practical understanding of how Proxies can be used to improve your own code. By the end of this session, you'll have a clear understanding of the capabilities of Javascript Proxies and how to apply them in your own projects.
Micro-Frontend & Me, a love story
I always thought Micro-frontends were stupid, unproven tech, and a fad. After speaking with industry leaders on my podcast about how they implemented it at scale, I realized that it might actually solve some issues we are having in our company. This talk focuses on the journey of implementing micro-frontends, from trade-offs, implementation, and choosing the right tools for the job.
🔥 Generative UI: The Future of Frontend
In the past few years, there have been a large number of generative AI use cases and adoption has grown increasingly. A large number of web and JS applications now leverage large language models (LLMs), vector search / RAG (Retrieval Augmented Generation) for building better user experiences. Now the tide is shifting, and we have entered the era of AI agents. This is a promising field which combines the different generative AI usecases and helps to simplify the process of interacting with Gen AI by essentially making agents act like humans and makes interactions feel more human like to solve problems in an automated fashion. In this workshop, while we give a quick over view of LLMs, RAG, we will dive deeper into concepts of AI Agents such has how one can leverage LLMs for specific tasks. How can we include tool calling / function calling in AI Agents. Thus showing how an AI Agent is a system that is capable of independently deciding which functions or data sources to used to efficiently execute users' orders. We will show how to develop an AI agent that runs purely on the browser vs running them using cloud. This workshop is for javaScript developers, AI enthusiasts, and tech innovators who want to integrate the use of AI within web applications to create intelligent and interactive app experiences. Table of contents: 1. Intro to the general landscape of LLMs, RAG 2. What are AI Agents 3. Core components of AI Agents 4. How to build your first AI Agent with JavaScript 5. Stack to build AI Agents in the browser with open source models
Bringing and Running AI Agents in the Browser
A large number of web and JS applications now leverage large language models (LLMs), vector search / RAG (Retrieval Augmented Generation) for building better user experiences. Now the tide is shifting, and we have entered the era of AI agents. This is a promising field which combines the different generative AI usecases and helps to simplify the process of interacting with Gen AI by essentially making agents act like humans and makes interactions feel more human like to solve problems in an automated fashion. In this talk, while we give a quick over view of LLMs, RAG, ,we will dive deeper into concepts of AI Agents such has how one can leverage LLMs for specific tasks. How can we include tool calling / function calling in AI Agents. Thus showing how an AI Agent is a system that is capable of independently deciding which functions or data sources to used to efficiently execute users' orders. We will show how to develop an AI agent that runs purely on the browser vs running it using cloud This talk is for JavaScript developers, AI enthusiasts, and tech innovators who want to integrate the use of AI within web applications to create intelligent and interactive app experiences.
WebAssembly + JavaScript: Partners, Not Competitors and why NOW is the right time to pay attention
WebAssembly (WASM) has been painted as a “JavaScript killer.” for many years. But in reality, I think it is opposite: WebAssembly and JavaScript are teammates that bring out the best in each other. What I have experienced that JavaScript continues to be the dynamic, flexible language of the web, while WebAssembly provides near-native performance for workloads that JS was never designed to handle. In this talk, we will break down this partnership with technical clarity and real-world examples, by concentrating on three points below: The execution model: How WASM compiles to a compact binary format, runs in a secure sandbox, and interacts with the JavaScript runtime through linear memory and import/export bindings. Interop in practice: Calling WASM from JavaScript using WebAssembly.instantiateStreaming, passing data between the two worlds, and handling memory efficiently (typed arrays, shared buffers). Performance hotspots: Why tasks like number crunching, media codecs, cryptography, and AI tensor ops can be offloaded to WASM, while keeping JS for orchestration, UI, and DOM. The best way to showcase this would be through following demos which will be interlinked within the talk: Demo 1 – Image Processing: Compressing and resizing images in WASM vs vanilla JS, showing 10x–100x speed improvements. Demo 2 – Data Crunching: Sorting millions of numbers in JS vs WASM, demonstrating raw computational gains. Demo 3 – AI in the Browser: Running a lightweight ML model (e.g., text embeddings or image recognition) with TensorFlow.js + WASM backend, proving why WASM is a game-changer for client-side AI. In addition we would do a tooling deep dive as an overview of wasm-bindgen (Rust), AssemblyScript, and TinyGo, with practical guidance on when to choose which. And we would also discuss beyond the browser which mean WASM on the server and edge, running in Node.js, Deno, and Cloudflare Workers with WASI (WebAssembly System Interface) unlocking file, network, and system-level capabilities. By the end, we will have, a clear technical mental model of how JS and WASM collaborate under the hood and awareness of current limitations (e.g., direct DOM access, debugging pain points) and how to work around them. Also a practical integration strategies we can adopt today without rewriting your app or learning a new low-level language. A vision of what iss next: polyglot apps where JavaScript orchestrates and WebAssembly accelerates. Takeaway of this talk is the future of the web is not JavaScript versus WebAssembly. It is JavaScript with WebAssembly — and with AI, edge computing, and high-performance apps becoming mainstream, now is the right time to embrace this partnership.
Building best-in-class apps, faster - using Browser DevTools and it’s super-powers!
This talk covers the most useful Browser DevTools capabilities you need to know - to maximize productivity, enhance DX and ensure a best-in-class customer experience. We will cover the most useful Browser DevTools capabilities, across key areas - General, Accessibility, CSS, Debugging, Testing and AI assistance. For each category, we will explore the following features: - General - Command palette | Console logging - group ; tabular ; timing | Current node access | Bundle analyzer - Accessibility - Emulation (vision impairment) | Color contrast - CSS - Visual editing (Grid ; flexbox ; angle & shadow editors) | Cascade layers - Debugging - 3rd party code | 3D view | HAR (HTTP Archive) | Animations - Testing - Local overrides | Emulation (Preferences) | Media queries - AI assistance - CSS / Layout | JS errors | Network requests | Code suggestions | MCP server
Building an AI-Powered Gym Assistant with JavaScript: When Your Code Has Better Form Than You Do
In an era where AI can write poetry and generate art, why not put it to work critiquing your workout form? This talk explores the practical implementation of an open-source gym assistant that leverages MediaPipe for pose estimation, React for an intuitive user interface, and Node.js for robust backend orchestration. We'll dive into: Real-time pose detection with MediaPipe: Implementing computer vision that's precise enough to catch your shortcuts (yes, half-reps count as shortcuts) Building a responsive React frontend: Creating an interface that provides constructive feedback without the judgmental tone of a personal trainer Architecting the Node.js backend: Managing real-time data streams and ensuring your server doesn't break a sweat, even if you do Overcoming real-world challenges: From handling variable lighting conditions to distinguishing between a push-up and someone dramatically collapsing This session combines modern web technologies with practical AI applications, demonstrating how JavaScript can extend beyond traditional web development into health and fitness domains. Whether you're passionate about fitness tech, curious about computer vision, or simply want to automate accountability, you'll walk away with actionable insights and a working knowledge of integrating AI into full-stack applications. Perfect for developers who believe technology should work as hard as we pretend to at the gym.
State of the Art: Transformers.js
Browser-based AI models are no longer a future dream — it’s here. In this lightning talk, I’ll show how Transformers.js lets you run cutting-edge transformer models directly in the browser, using only JavaScript. We’ll look at how WebGPU/WebAssembly made this possible, practical use cases for real-world apps, and why client-side AI will reshape the frontend ecosystem. Keep you curiousity levels at peak as we gonna add LLM capabilities to your UI without any server dependency
Would you like an opportunity to speak at the CityJS event?
Send us a submission by 30th Dec