Monthly Meeting (December 2025)

Data Science Meetup – 10th Anniversary Celebration!
As we wrap up 2025, we're thrilled to celebrate a special milestone: Completion of 10 years of our Data Science meetup group!
Join us for our next session:
Date: Saturday, December 13th
Time: 3:00 PM EST
Location: Yoga Studio at Radha Gopinath Temple
7 Kilmer Court, Edison, NJ
Agenda:
3:00 – 3:30 - The journey so far and Review of Trends in AI – Organizer team
3:30 – 4:15 - Exploring MCP – Saravanan
4:15 – 5:00 - Networking , Coffee & Snacks
We look forward to seeing you there as we reflect on a decade of learning, sharing, and growing together.
Send email to spn2113@gmail.com to request for online meeting link if you need one.
For those who missed last month , below is a summary of the session
AI Agent Building with Python
The meeting began with a discussion about job opportunities, including an AI architect position at a company in Parsippany, which requires 4 days a week and 2-3 years of experience with AI. The group then transitioned to a presentation by Gus about building AI agents using Python. Gus explained that the session would mix context slides with hands-on labs, using Google Colab for the demonstrations. He noted that attendees would learn about using GenAI Quick Start from Google, which is free and accessible either locally or through Colab notebooks. The conversation ended with Gus introducing himself and sharing a quiz question about Python data analysis tools, which Saravanan and Purush answered correctly.
AI Implementation and ROI Strategies
Gus led a discussion on the current state of AI and machine learning, emphasizing the need to focus on specific, actionable workflows rather than general AI tools. He highlighted the challenges businesses face in understanding and implementing AI effectively, citing examples like Walmart's dual approach to AI implementation. Gus also discussed the importance of ROI in AI investments and criticized companies that overhype their AI capabilities. He concluded by advocating for learning to code and build software, predicting an increase in software development as AI tools become more accessible. The session ended with Gus preparing to demonstrate a practical example of using AI in a specific business context.
AI Tools in Software Development
The meeting focused on the adoption and impact of AI tools in software development, particularly Copilot, and the challenges faced in integrating AI into existing workflows. Saravanan highlighted that while AI tools like Copilot can significantly improve productivity, there is a concern about code quality due to potential complacency among developers. He also noted that only a small percentage of developers are actively using these tools, despite the potential benefits. The discussion touched on the need for proper training and a supportive environment to encourage tool adoption. Additionally, Gus introduced a tutorial on building agents using Google's GenAI, demonstrating how tools like Google Search can enhance the capabilities of large language models. The conversation ended with a discussion on the importance of building reliable, specific tools to improve the predictability and effectiveness of AI in various tasks.
AI Agents and Enterprise Integration
Gus discussed the importance of programming agents to take discrete actions and emphasized that most value initially comes from strategic domain expertise rather than AI alone. He used Microsoft's example of optimizing customer onboarding from 230 to 40 steps, noting that the reduction was primarily achieved through human reasoning rather than AI. Gus explained the concept of agents interacting with enterprise data securely and governed through platforms like Snowflake and Databricks, allowing for text-based data exploration and analysis. He concluded by introducing the concept of agent-to-agent communication protocols, contrasting internal tool integration (MCP) with external agent interactions, and highlighted the growing trend of specialized AI services akin to Web 2.0, suggesting a future where different AI agents will communicate horizontally across platforms.
AI Agents and Future Roles
Gus presented on the concept of agents and their role in the future of AI, discussing the Model Context Protocol (MCP) and agent-to-agent communication. He explained how agents can connect to various tools and code blocks, and demonstrated a simple example using Python. The group discussed the potential for new job roles, such as agent architects, and the need for better data governance in AI systems. There was also a discussion about the current ROI of AI investments, with some skepticism about the long-term viability of the business model.