Karthik Nadig

About Karthik Nadig

I build AI-powered developer platforms used by tens of millions of engineers.I currently lead engineering areas in Visual Studio Code focused on AI model evaluation, developer infrastructure, and intelligent developer workflows. My work sits at the intersection of frontier AI models and real-world software engineering workflows.I lead teams responsible for multiple core capabilities inside VS Code including:• AI model evaluation systems used to validate and improve coding models• Terminal infrastructure used by both developers and AI agents• Integrated browser capabilities enabling agents to interact with real applications• The Python developer ecosystem used by ~14M developers monthlyA significant portion of my work involves collaborating directly with frontier AI model providers to improve coding model performance in real developer workflows. I design and operate advanced evaluation frameworks that benchmark models across latency, tool usage, trajectory behavior, and code quality.These systems evaluate 25+ models across benchmark suites built specifically for real-world developer workflows. Several model providers have described the benchmark methodology as one of the most advanced real-world evaluation systems they have encountered.In addition to leading internal engineering efforts, I regularly collaborate with teams across the company building AI-powered developer experiences and frequently advise other organizations integrating modern AI models into engineering workflows.Earlier in my career I built core components of the Python developer ecosystem in VS Code, including the modern Python debugger used across multiple developer environments.My work combines three areas I care deeply about:• AI systems and model evaluation• developer platforms and programming tools• building infrastructure used at global scaleI’m particularly interested in advancing the next generation of AI-powered developer tools and intelligent software engineering systems.

Experience

1. Principal Software Engineering Manager

   Microsoft

   Sep 2024 — Present · Redmond, WA, US

   I lead engineering areas in Visual Studio Code focused on AI-driven developer experiences and platform infrastructure.My teams work on multiple core systems inside VS Code including:• AI model evaluation systems used to benchmark and validate frontier coding models• Terminal infrastructure used by both developers and AI agents• Integrated browser capabilities enabling agents to interact with real applications• The Python developer ecosystem used by ~14M developersA key part of my work involves collaborating directly with frontier AI model teams to improve real-world performance in developer workflows. I design and operate evaluation frameworks used to benchmark coding models across dimensions including latency, tool usage, trajectory analysis, and code quality.These systems currently evaluate 25+ models across benchmark suites built specifically for developer workflows.I also work closely with multiple internal teams building AI-powered developer tools and frequently advise organizations integrating modern coding models into engineering systems.

Education

• Institute Of Language Studies

  German Language Certificate, German

  2009 — 2010

• The University of Georgia

  MS, Artificial Intelligence

  2010 — 2012

• Vidya Vardhaka College of Engineering

  B.E, Electronics & Commnication

  2003 — 2007

Skills

• Algorithms
• Javascript
• Xaml
• C++/Cx
• Eclipse
• Neural Networks
• Software Development
• Cpp
• Python
• Software Engineering
• Image Processing
• Robotics
• Object Oriented Design
• Visual Studio
• Php
• Computer Vision
• Natural Language Processing
• Computer Science
• Pattern Recognition
• Mysql
• Distributed Systems
• Programming
• Win32 Api
• Git
• Software Design
• Silverlight
• Xml
• Rest
• Cx
• C#
• Oop
• Winrt
• Wpf
• Machine Learning
• Artificial Intelligence
• Wcf
• Perl
• Windows 8
• Linux
• Com