So, you want to build a Generative AI agent for customer conversations in production. However, bringing such a system into production comes with key challenges, such as mitigating hallucinations, evaluating performance pre-deployment, choosing the right model, and balancing accuracy, scalability, and latency. Thoughtful design decisions are crucial to ensuring reliability, user experience, and cost efficiency.
At Apartment List, we are building Generative AI assistants tailored for the rental housing sector. Just months after ChatGPT transformed AI-driven conversations, we redesigned our conversational engine to be powered by Generative AI – deploying our first production-ready AI system over a year and a half ago. In this talk, I’ll share key insights from our ongoing journey – what worked, what didn’t, and the critical factors for scaling Generative AI successfully. We’ll explore streaming vs. non-streaming AI agents, key design trade-offs, best practices for evaluation and monitoring, and how to optimize AI for real-time interactions.
Whether you’re an NLP researcher, a data scientist working with LLMs, or an AI enthusiast, this talk will provide practical insights and real-world strategies for successfully building and deploying Generative AI applications at scale.
I am a Senior Data Scientist at Apartment List, where I work on developing virtual AI assistants for the rental housing industry using generative AI and machine learning. I lead projects from research to production, building intelligent, data-driven solutions that optimize communication and user engagement. I hold both a BSc and MSc in Computer Science from Ben-Gurion University.
My passion for data science lies in leveraging AI to solve complex real-world challenges. Before my current role, I was a Principal Data Scientist at Oracle Advertising, designing algorithms that process billions of events daily to generate insights for advertisers and publishers. Prior to that, at IBM Research, I applied Al to healthcare, contributing to multiple research projects on breast cancer detection, which led to published studies.
My expertise in data science is rooted in both industry and academia. During my MSc at Ben-Gurion University, I conducted research integrating machine learning with biological data and published a paper proposing a novel machine learning-based approach for classifying pathogenic bacteria. During my undergraduate studies in Computer Science, I specialized in Bioinformatics and was awarded the prestigious Schulich Leaders Scholarship.
I am deeply passionate about the intersection of AI, data science, and real-world impact. Throughout my career, I have worked on a diverse range of data-driven applications, from large-scale data processing to predictive modeling and generative AI. I am excited about the future of data science and continuously explore innovative ways to apply data-driven insights to meaningful problems.
8:45 | Reception |
---|---|
9:30 | Opening remarks by WiDS TLV ambassadors |
9:45 | Dr. Mor Geva , Tel Aviv University: “MRI for Large Language Models: Mechanistic Interpretability from Neurons to Attention Heads” |
10:15 |
Panel: “Pioneering Progress: a strategic look at the GenAI revolution and the new role of data scientists“ Shani Gershtein, Melingo Mirit Elyada Bar, Intuit Dr. Asi Messica, Lightricks Moderated by Nitzan Gado, Intuit |
10:45 | Poster pitches |
10:55 | Break |
11:10 | Lightning talks session |
12:30 | Lunch & poster session |
---|---|
13:30 | Roundtable session & poster session |
14:30 | Roundtable closing |
14:40 | Shunit Agmon, Technion: “Bridging the Gender Gap in Clinical AI: Temporal Adaptation with TeDi-BERT” |
15:00 | Shaked Naor Hoffmann, Apartment List: “Building Generative AI Agents for Production: Turning Ideas into Real-World Applications” |
15:20 | Closing remarks |
15:30 | The end |
WiDS Tel Aviv is an independent event that is organized by Intuit’s WiDS TLV ambassadors as part of the annual WiDS Worldwide conference, the WiDS Datathon, and an estimated 200 WiDS Regional Events worldwide. Everyone is invited to attend all WiDS conference and WiDS Datathon Workshop events which feature outstanding women doing outstanding work.
© 2018-2024 WiDS TLV – Intuit. All rights reserved.
Scotty – By Nir Azoulay
Design: Sharon Geva