What a Gen AI Scientist at Property Software Company Wishes They Had Known Before Entering the AI Industry
Gagandeep, a Gen AI Scientist, wishes they had known the crucial role of "domain knowledge" and the "iterative nature of AI development," highlighting the real-world challenges of data preparation and model refinement beyond textbook examples. The ethical implications of AI, building models that are "fair, transparent, and beneficial to society," are also key learnings from their experience.
Data Analysis, Ethical Implications, Iterative Development, Domain Knowledge, Problem-Solving
Advizer Information
Name
Job Title
Company
Undergrad
Grad Programs
Majors
Industries
Job Functions
Traits
Gagandeep Singh
Gen AI Scientist
Property Software Company
Uttar Pradesh Technical University
Arizona State University (ASU) - W. P. Carey
Computer Science
Real Estate, Technology
Data and Analytics
None Applicable
Video Highlights
1. The importance of domain knowledge in AI is crucial for building effective models and interpreting results meaningfully. Real-world AI development requires understanding the business context and domain-specific challenges, not just mastering algorithms and coding skills.
2. AI development is iterative. Unlike school projects with clean datasets, real-world projects involve significant time spent on data preparation, feature engineering, and model refinement based on feedback.
3. Ethical implications of AI are paramount. Building AI systems requires considering their societal impact and potential biases to ensure fairness, transparency, and societal benefit. It's not just about performance, but responsible development.
Transcript
What have you learned about this role that you wish someone had told you before you entered the industry?
One thing I wish I had known before entering the field is the importance of domain knowledge. When I was starting out, I was very focused on mastering algorithms and improving my coding skills. While these are crucial, I've learned that understanding the business context and domain-specific challenges is equally important.
For instance, when I worked on insurance enrollment at Conduent, I realized that having a deep understanding of the insurance industry, especially health and welfare insurance, significantly improved my ability to build effective models and interpret results meaningfully.
Another aspect I wish I had grasped earlier is the iterative nature of AI development. In school, we often work on well-defined problems with clean datasets, but that's not true in the real world. Most of our time is spent on data preparation, feature engineering, and redefining our models based on feedback. It's really a straightforward process from problem to solution.
Lastly, I wish I had understood the ethical implications of AI earlier in my career. As AI systems become more powerful and pervasive, considering their societal impact and potential bias is crucial. It's not just about building a model that performs well; it's about building one that is fair, transparent, and beneficial to society.
Advizer Personal Links