How can we use Generative AI to analyze transcripts or notes of client conversations and automatically extract actionable insights, such as client requests and action items?
Learn more about the challenge
Access the data
Q&A Session for hackers
🏗️ 🚧 to be defined 🏗️ 🚧
Problem Client conversations contain key pieces of information, but manually analyzing transcripts or meeting notes to extract actionable insights is time-consuming and prone to error. For this challenge, we invite participants to develop a solution that leverages Gen AI to automatically analyze conversation transcripts and extract key pieces of information.
The goal is to empower client advisors with concise, accurate, and actionable insights, helping them save time and better serve their clients. The solution should be capable of:
- Extracting specific information, such as client requests, and agreed-upon actions.
- Ensuring accuracy and relevance, even in the presence of noisy or incomplete data.
Objective
- Information Extraction: Accurately identify and extract key elements, including action items or follow-ups.
- Accuracy and Relevance: Ensure that the extracted information is highly accurate and relevant to the context of the conversation.
- Robustness: Handle noisy or incomplete data gracefully, producing reliable outputs even in challenging conditions.
- Cost-Accuracy Comparison: Explore and compare different SLMs and LLMs, focusing on their performance and cost-efficiency.
Support for Hackers
- You can expect in-person and online support from our data scientists and senior data scientists.
Technical Preferences
- “UBS friendly” tech stack would be preferred (Azure, Python, Gitlab).
Why hack? This challenge offers a unique opportunity to work on cutting-edge Generative AI technologies in a domain with real-world impact. By participating, you will:
- Contribute to shaping the future of client interaction in banking.
- Gain valuable experience in NLP, information extraction, and working with financial-domain challenges.
- Explore the trade-offs between performance and cost in AI model selection.
About the Challenge Partner
UBS is a leading and truly global wealth manager and the leading universal bank in Switzerland. It also provides diversified asset management solutions and focused investment banking capabilities. UBS manages over 6 trillion dollars of invested assets (as at 31 Dec 2024) and helps clients achieve their financial goals through personalized advice, solutions, and products.
Building on more than a decade of experience in this field, they have put AI as a key driver of their strategy helping to reshape their business to benefit clients and employees
As a main partner of Swiss {ai} Weeks, UBS brings deep insight into how AI is reshaping the financial sector and global economy.