CST 349: Computer Science Proseminar

Course Description

The Proseminar course helps students prepare for life after college by connecting their major to broader personal and professional goals. In this course, students critically analyze information from diverse sources, including academic and business literature and conversations with industry mentors. Students also develop crucial problem-solving, communication, and teamwork skills throughout the course.

This analysis and skill development guide students in crafting their personal value proposition and a plan for developing a professional portfolio.

Prerequisite: Junior Standing Required

Typically Offered: Fall, Spring

Units: 2

Industry Expert Interview Report

For this assignment, I interviewed a Senior Data Architect and Engineer who has worked extensively within the Databricks ecosystem for over ten years. Their experience spans enterprise data platform architecture, large-scale distributed systems, cloud infrastructure design, and advanced analytics implementation.

Professional Background

The professional began working with Apache Spark during its early adoption phase and transitioned into building large-scale data engineering solutions using Databricks as it matured into an enterprise analytics platform. Over the past decade, they have designed data lakes, implemented Delta Lake architectures, optimized ETL pipelines, and led cross-functional engineering teams delivering cloud-native data solutions.

Key Insights from the Interview

Industry Perspective

The interview emphasized that the data engineering field has shifted from traditional database administration toward scalable cloud-native architectures. Databricks, in particular, was described as a platform that integrates engineering, analytics, and machine learning into a unified workflow. The expert highlighted the importance of understanding not just how to use tools, but why certain architectural decisions are made.

Personal Reflection

This conversation reinforced the importance of building strong foundational knowledge in computer science while remaining adaptable to evolving technologies. Hearing firsthand how large-scale distributed systems are implemented in production environments clarified the skills I need to prioritize: system design, data structures, cloud computing, and collaborative problem-solving.

The interview also helped me better define my professional direction by demonstrating how technical depth combined with communication and leadership skills can create long-term career growth. This experience strengthened my understanding of how to position myself within the technology industry and how to continue building a portfolio aligned with modern data engineering practices.

Load Balancing: How Websites Stay Online During Traffic Spikes

For our CST 300 Final Research Project, our team explored how modern websites stay online when millions of users access them at the same time. Our focus was load balancing, a core distributed systems concept that spreads incoming traffic across multiple servers instead of overwhelming a single machine.

We examined real-world scale and impact, including traffic surges during product launches, viral events, online sales, and streaming releases. Our research connected technical infrastructure to business outcomes, showing how downtime can translate into major financial loss and degraded user experience.

Key Concepts Covered

  • Single-server bottlenecks
  • Traffic spikes and nonlinear demand
  • Fault tolerance and high availability
  • Horizontal vs. vertical scaling
  • Routing algorithms (Round Robin, Least Connections, IP Hash)
  • Cloud-based load balancing systems (AWS, Azure, Google Cloud)

At its core, load balancing acts as an intelligent traffic director. Instead of allowing one server to become overloaded, requests are distributed across many machines to ensure performance, reliability, and uptime.

Development Process: Fun Version (2–5 Minute Video)

For the short general-audience version, our goal was to explain load balancing in a clear, engaging way without technical jargon. We intentionally designed this version for a middle-school-level audience while maintaining technical accuracy.

1. Start With a Relatable Problem

We opened with everyday overload scenarios:

  • Black Friday store crowds
  • Concert ticket websites crashing
  • Traffic jams during rush hour

These examples illustrated the core idea: when too many people try to use one resource at the same time, systems fail.

2. Introduce the Bottleneck Concept

We visually demonstrated how a single limited resource creates congestion:

  • One door with hundreds of people
  • One cashier with a massive line
  • One server handling thousands of users

This established the transition: the system is not broken — it is overloaded.

3. Use the Grocery Store Analogy

We then introduced multiple checkout lanes to show how distributing work improves flow:

  • One cashier → long lines
  • Multiple cashiers → shorter, faster lines
  • Store employee directing customers → load balancer

This directly mapped to the technical concept:
Users → Load Balancer → Multiple Servers

4. Connect to Everyday Technology

Finally, we showed how load balancing improves daily life:

  • YouTube handling millions of viewers
  • Online gaming preventing lag
  • Retail websites staying online during peak sales

Collaboration & Workflow

Each team member was responsible for a defined section of the outline, which ensured structure and clarity. We divided the content into problem → analogy → explanation → real-world impact to maintain logical progression.

We designed visuals first, then aligned scripts to the visuals. We iterated on slide structure, transitions, and tone to ensure strong audience engagement and alignment with the Online Video Presentation Rubric.

Overall Takeaway

This project demonstrates that load balancing is foundational to modern internet infrastructure. By creating both a professional and a general-audience version, we translated complex distributed systems concepts into accessible explanations while maintaining technical depth and accuracy.

Additional Work Completed in CST 349