How Devs Can Maximize Their Productivity with AI Agents in 2025

This conference talk by Tejas Kumar explores how developers can offload work to large language models and automate routine tasks to enhance daily productivity. Learn how LLMs can call functions and discover practical applications through demonstrations of AI agent tools. The presentation covers building custom AI agents with Langflow, performing currency conversions, integrating Composio for enhanced productivity, utilizing the Model Context Protocol with GitHub, and searching and fixing code issues automatically. The 25-minute talk includes detailed walkthroughs of setting up these tools, conducting experiments to find free schedule slots, exploring different AI models, and troubleshooting common issues. Perfect for developers looking to leverage AI agents to eliminate busywork and focus on higher-value tasks in their development workflow.

Syllabus

00:00 Introduction to AI Agents
03:14 Agent Components and Tool Usage
05:12 Building Your Own AI Agent with Langflow
07:00 Converting Euros to Indian Rupees
08:36 Enhancing AI Agent Productivity with Composio
10:24 Setting up Composio with Langflow
11:37 Using Tools and Checking Schedule
12:45 Conducting Experiments and Finding Free Slots
14:33 Langflow and Model Context Protocol
16:33 Using Model Context Protocol with GitHub
18:41 Exploring MCP Server Capabilities
20:32 Exploring Different Models and Fixing Issues
22:17 Using MCP to Search and Fix Code Issues on GitHub
24:06 Concluding Remarks and Next Steps