Homepage
Sign inGet started
Data Science at Microsoft
  • Methods
  • Visualization
  • Engineering
  • Organization
  • Development
  • People
  • All
Methods
A practical guide to building enterprise-grade RL agents under strict privacy constraints

A practical guide to building enterprise-grade RL agents under strict privacy constraints

This article highlights an approach we developed for improving an enterprise agent’s ability to assist users effectively. At Microsoft…
Go to the profile of Aditya Challapally
Aditya Challapally
May 26
A primer for understanding quantum computing in 2025

A primer for understanding quantum computing in 2025

In May 1981, MIT held a conference on the physics of computation. Organized in conjunction with IBM at MIT’s Endicott House in Dedham…
Go to the profile of Darshkodwani
Darshkodwani
May 21
Leveraging Mean Opinion Score for quality assessment in computer vision

Leveraging Mean Opinion Score for quality assessment in computer vision

In an era in which Artificial Intelligence is shaping how we see, hear, and interact with digital content, understanding human perception…
Go to the profile of Darina Bal Roitshtain
Darina Bal Roitshtain
May 19
Leveraging Explainable AI to proactively mitigate bias in automated decision systems

Leveraging Explainable AI to proactively mitigate bias in automated decision systems

We often hear that AI “hallucinates” in certain scenarios or that it can’t handle complex tasks because it “doesn’t understand” the…
Go to the profile of Sajal Mittal
Sajal Mittal
May 14
The stability-plasticity dilemma in AI

The stability-plasticity dilemma in AI

In the continuing odyssey of engineering truly sentient systems, one of the most formidable yet exhilarating frontiers is the intricate…
Go to the profile of Aparana Gupta
Aparana Gupta
May 12
AI agents: The digital sidekicks we never knew we needed!

AI agents: The digital sidekicks we never knew we needed!

The landscape of Artificial Intelligence is rapidly evolving, with significant advancements transforming the way we interact with…
Go to the profile of Eva Jurado Cortés
Eva Jurado Cortés
May 7
How to improve mathematical skills with AutoGen agents

How to improve mathematical skills with AutoGen agents

Large Language Models (LLMs) like GPT-4 have revolutionized the field of natural language processing, enabling a wide range of applications…
Go to the profile of Marco Zatta
Marco Zatta
May 5
Stop the spill: The blueprint for eradicating data leakage

Stop the spill: The blueprint for eradicating data leakage

In the high-stakes arena of Machine Learning, every data point is cherished as a beacon of untapped insight and potential. Yet, hidden in…
Go to the profile of Aparana Gupta
Aparana Gupta
Apr 30
Empowering field engineers with SLM-based AI copilots in disconnected mode

Empowering field engineers with SLM-based AI copilots in disconnected mode

Welcome to an era of Small Language Models
Go to the profile of Chandan Srivastava
Chandan Srivastava
Apr 28
SLMs on edge: A classifier approach for function calling

SLMs on edge: A classifier approach for function calling

The game changed with transformer-based architectures such as Google’s BERT [3] in 2018 and OpenAI’s GPT [4], showcasing unprecedented…
Go to the profile of Darshkodwani
Darshkodwani
Apr 23
Safety and security in agentic systems

Safety and security in agentic systems

Safety and security in agentic AI systems are critical due to their autonomous nature and evolving capabilities
Go to the profile of Davidjonietz
Davidjonietz
Apr 21
Continuous improvement with agentic AI: Conducting a virtual gemba walk

Continuous improvement with agentic AI: Conducting a virtual gemba walk

Leveraging AI-driven multi-agent systems (DeepSeek, AutoGen, Ollama) to revolutionize process optimization, eliminate inefficiencies, and…
Go to the profile of Mehul Jain
Mehul Jain
Apr 16
Comprehensive evaluation framework for LLM-generated content

Comprehensive evaluation framework for LLM-generated content

By Juhi Singh, Bonnie Ao, and Sebastian Antinome
Go to the profile of Juhi Singh
Juhi Singh
Apr 14
Data currency: Gain the temporal edge in AI

Data currency: Gain the temporal edge in AI

In today’s hyper-connected digital era, where every millisecond can redefine markets and reshape industries, the concept of data currency…
Go to the profile of Aparana Gupta
Aparana Gupta
Apr 7
Optimizing prompt refinement: Algorithmic strategies for LLM-driven classification

Optimizing prompt refinement: Algorithmic strategies for LLM-driven classification

By Juhi Singh, Bonnie Ao, and Sebastian Antinome
Go to the profile of Juhi Singh
Juhi Singh
Mar 31
About Data Science at MicrosoftLatest StoriesArchiveAbout MediumTermsPrivacyTeams