Patrice Bechard

Thinking...

<think>
Alright, time to put together a bio that captures both my research focus and my broader background. First, I should clearly frame my current work at ServiceNow. The key theme is AI agents for enterprise automation. My research is not just about applying language models, but about building systems that can reliably interact with enterprise platforms. That includes designing realistic environments and benchmarks, studying how agents operate through different interfaces (web interfaces, APIs, documentation), and evaluating how well they perform on complex enterprise workflows. It’s important to highlight that this work sits at the intersection of AI research and real-world systems. Next, I should summarize the applied work I’ve done at ServiceNow. This includes projects on text-to-workflow generation, image-to-workflow generation, and AI systems that automate operational tasks such as IT issue resolution. These projects show how machine learning can be integrated directly into enterprise platforms to automate real business processes. Then I should cover my earlier experience at Intact Financial. The focus there was applied NLP for insurance applications: conversational agents for call centers, large-scale analysis of call transcripts, and deep learning models applied to telematics data. This part of my career highlights experience building machine learning systems that operate on large, real-world datasets and support business decision-making. After that, I can briefly mention my academic background. My Master’s work at MILA in machine learning and my undergraduate degree in physics provided the theoretical foundation for my current research. Teaching assistant work in relativity and astrophysics research also helped develop strong analytical and communication skills. Finally, I should include a short personal section. Running, hiking, and exploring restaurants and wine bars are good ways to show personality and interests outside of research. I also want to mention languages and curiosity for learning new things. The goal is to present a coherent narrative: research on AI agents and automation today, applied machine learning experience in industry, and a strong scientific background.
</think>

# 👋 About Me

I am an Applied Research Scientist at [ServiceNow](https://www.servicenow.com/), where I work on **AI agents for enterprise automation**. My research studies how large language models can reliably interact with complex enterprise systems to execute real-world workflows. I design **evaluation environments and benchmarks for enterprise tasks**, and investigate how different classes of agents—from web-based agents operating through graphical interfaces to programmatic agents interacting directly with APIs and documentation—can solve challenging automation problems. My goal is to develop **reliable agent architectures and evaluation frameworks** that enable practical AI automation in real-world enterprise environments.

At ServiceNow, I have contributed to several initiatives applying machine learning and language models to enterprise automation. This includes research on **text-to-workflow generation** and **image-to-workflow generation** for [ServiceNow Flow Designer](https://www.servicenow.com/products/platform-flow-designer.html), as well as systems that automate IT service management tasks such as [**issue auto-resolution**](https://www.servicenow.com/docs/bundle/washingtondc-servicenow-platform/page/administer/virtual-agent/concept/auto-resolution-va.html). My work has involved building datasets and benchmarks, training domain-adapted models, and deploying AI capabilities that help organizations automate operational workflows more efficiently.

Before joining ServiceNow, I was a Data Scientist at [Intact Financial Corporation](https://www.intact.com/en/), the largest property and casualty insurer in Canada. There, I worked on natural language processing and conversational AI systems for insurance applications. My work included analyzing call center conversations, modeling customer intents, assisting support agents during interactions, and applying deep learning techniques to telematics and insurance data.

## 🎓🤖🪐 Education

I hold a Master’s degree in Machine Learning from [MILA - Quebec AI Institute](https://mila.quebec/en/), and a Bachelor’s degree in Physics from [Université de Montréal](https://phys.umontreal.ca/english/home/). During my studies, I worked as a teaching assistant for courses in special relativity and briefly contributed to astrophysics research at the [Centre de Recherche en Astrophysique du Québec (CRAQ)](https://craq-astro.ca/?lang=en).

## 🏃🍕🍷 Hobbies

Outside of research, I enjoy long-distance running, hiking, and exploring new cities through their food scenes—especially natural wine bars, craft breweries, and great local restaurants. I’m fluent in French and English, enjoy practicing Spanish, and generally enjoy learning new things, whether in AI, cooking, or the outdoors.

You can find a list of my research publications on [Google Scholar](https://scholar.google.com/citations?user=O-YzTloAAAAJ&hl=en).