Interview with – Arthur Jansen
summary
Company sells B2B business signals and builds a new lead product that scores prospects from website content instead of just annual financials, letting models pick up dynamic hires, product pages, and signals of change. Pipeline processes several million prospects and uses CNN for images, representation learning on the company graph, and fast NLP models in production because Transformers are too slow at that scale. Early results show higher lead recall and more timely signals than static financial feeds.
AI-written summary · about
chapters (8)
- 0:00Start
- 0:20Can you explain to us what D&B do?
- 0:55You are working as a Data Science Specialist at Dun & Bradstreet. Could you elaborate on your role a bit? How much is data engineering and data science essential in your everyday operations?
- 1:31Can you elaborate how D&B developed a new way of generating leads by integrating online data and using multiple Machine Learning models and how it works?
- 2:15As you used multiple Machine Learning models in this project, which ML models perform better in which areas?
- 3:30What advantages do you see in these sorts of conferences, and what have been some of your biggest takeaways thus far?
- 4:06Is there anything a bit more beside data science specialist?
- 5:30Could you expand a bit on how you started of in data science?
related talks
21mBuilding AI that Searches and Sells
11mGenesis Startup Competition Pitch: SparkBeyond
08mQ n A with Dr Arun Verma
47mAI Pitch: Beyond the Hype
29mAI-Driven Business Success: Measurable Results and Revolutionized Sales
35mUsing AI to Fuel Digital Transformation & Growth
28mBeyond the Balance Sheet: Transactional Data as a Window to Company Health
23mDelivering business success with AI