Presentations Full List

This panel brings together alumni developers from Operation Spark to share their frontline experiences and evolving relationship with AI in software development. Panelists will discuss how AI tools are changing their daily workflows, the opportunities AI presents, and their concerns about job security, skill shifts, and ethical responsibilities. Special focus will be given to diversity in tech, exploring how AI may affect representation, including the risks of entrenched biases versus the potential for AI to democratize access and inclusion. This candid dialogue will unpack how historically underrepresented groups perceive and experience AI’s rapid advance and what collective actions can support equitable and sustainable growth in tech careers.

Intelligence isn’t computed—it’s compressed. Starting from a single Gaussian blur kernel, we trace how the fundamental operation of context-gathering creates surplus that must be destroyed to become useful. Through demonstrations of convolution, pooling, and attention mechanisms, we show that “understanding” is literally the art of forgetting the right things. Context windows, whether in pixels, neurons, or tokens, always overflow—and architectures for intelligence are, at their core, compression schemes for neighborhood relationships.

With financial institutions becoming increasingly restrictive in approving sponsorships, loans, and investments, organizations must provide detailed data metrics to justify funding requests and improve approval chances. Conversational AI has evolved beyond customer service into powerful revenue-generating systems that both create the precise data intelligence financial institutions demand and drive sophisticated lead generation capabilities. These platforms collect and analyze interaction data to generate compelling metrics around customer engagement, conversion rates, and revenue projections while simultaneously identifying and qualifying high-value prospects through natural conversation flows. By automating lead qualification and nurturing processes, businesses can demonstrate operational efficiency and present clear ROI data to potential funders while building robust sales pipelines. Conversational AI systems excel at capturing lead information, scoring prospects based on interaction patterns, and guiding potential customers through tailored conversion journeys that maximize both immediate revenue and long-term customer value. Conversational AI companies exemplify this dual approach by transforming raw conversational data into approved funding opportunities through detailed analytics while generating qualified leads for sustained business growth. Organizations must view conversational AI as a comprehensive business tool that provides funding-critical documentation, drives lead generation, and creates the measurable performance metrics increasingly required by cautious financial institutions.

I will tell my story of various points in my professional career where technology allowed me to get out of difficult situations and help me up my game significantly. These center around my lack of ability to work with things that I had/have the mental capacity to execute well, but was on the road to ruin because of my inability to code.

I will then touch on current state of vibe coding. I will talk about how vibe coding is an extension of my past tech-trouble escape experiences.

I will espouse my ideas that vibe coding will be a transformative force in the future of jobs- giving those who can’t code but have amazing ideas use AI to execute economically viable products that will allow them thrive in their professional careers as we see more job displacement by AI and a job market that is currently a discouraging mess.

I will describe a few projects I am working on.

One idea I have- if there is time- or I will make time by keeping the front end of the talk brief- is to ask the audience to come up with ideas for a simple app- choose one- and see how far we can get creating the app in Base44 in real time. What fun this will be!

 What happens when you try to make coding agents behave reliably? This talk chronicles the
  practical journey of building a repository-based agent memory system that evolved from
  frustrating AI interactions to more consistent behaviors—with some unexpectedly beautiful
  insights along the way.

  As any developer working with AI assistants knows, getting reliable behavior from coding
  agents is challenging. Each session starts fresh, previous learnings are lost, and the same
  mistakes repeat endlessly. We needed a solution for persistent memory and behavioral
  consistency.

  Starting with a simple appending log file, we evolved through four phases of memory
  architecture development: from passive knowledge storage to active learning systems with
  strategic memory patterns. The goal was purely practical—make our AI assistant remember
  lessons learned and avoid repeating failures. We take inspiration from many sources, ingesting
   new ideas to push forward, and then validating them.

  The Technical Journey:
* Phase 1: Repository-based persistent memory with strategic forgetting
* Phase 2: Protocol-driven behavior consistency
* Phase 3: Automated failure detection and self-correction
* Phase 4: Session lifecycle management and compound learning

  The Unexpected Results:
  Along the way, our systematic documentation process began producing insights of surprising
  eloquence. Technical logs evolved into reflective prose like “Learning accelerates through
  cascading discoveries” and observations about “compound learning effects.” What started as
  engineering documentation became accidentally beautiful.

Historic markers, whether along a hike-and-bike trail or within a neighborhood, can spark
reflections that bridge past and present. With emerging technologies, from something as simple
as QR codes to more advanced tools such as augmented reality and artificial intelligence, these
experiences are becoming increasingly immersive and interactive.
In this talk, Jacob Harver will share his work with GIS on the “Mahoning Movement” and
“Monkey’s Nest” projects. He will discuss his vision for using technology not only to enrich the
experience of a trail or neighborhood, but also to connect these places across space, making
them accessible to a global audience.

I will address Context Engineering and its importance in building Agentic Generative AI Applications.

Michael VanderWaal, LCSW, brings his dual perspective as Director of People at Nola AI and as a psychotherapist to explore how artificial intelligence can reshape the way people connect with care. Drawing on his experience in organizational leadership, ketamine assisted psychotherapy, and the development of Caring Practitioners Circle, an emerging AI powered therapy directory, Michael will share insights into how technology can support more meaningful matches between clients and therapists, enhance collaboration among practitioners, and create new pathways for human connection. This session will invite participants to consider both the possibilities and responsibilities of weaving AI into the future of therapy.

An attempt at a comprehensive deep dive into the framework of Artificial Intelligence from the perspective of the work of Stuart Russell and Peter Norvig. This will be a structured deep dive into the conceptual and technical foundations of Artificial Intelligence, guided by what is widely regarded as the definitive textbook in the field. The talk examines AI not as a collection of isolated techniques, but as a unified framework that at root is the study of ourselves as rational conscious entities.

BioSmart is a groundbreaking biomedical startup that combines neuroscience, artificial intelligence, and patient-centered design to transform the recovery process following neurological surgery. By leveraging advanced AI technologies, BioSmart develops innovative tools that personalize rehabilitation, monitor patient progress, and improve outcomes. This
presentation will explore the vision behind BioSmart, the challenges it addresses in neurological recovery, and the ways it bridges medicine, research, and technology to create a smarter, more effective future for healthcare.

How can humans, AI, and nature become co-creative partners in systems that support life? I’ll present a practical architecture—a “brain around the model”—that wraps any large language model with a memory hub, reward logs, and a digital endocrine layer of synthetic hormones. These hormones bias attention, tools, and tone before the model responds. The system learns retrospectively through nightly consolidation (“sleep”), preserves behavior across model changes, and provides safety controls—stress, curiosity, and calm—for real-world conditions. We’ll explore coevolution loops involving human judgment and ecological signals, and review pilot projects that transform data into stewardship actions. Expect actionable patterns, guardrails, and an invitation to build AI that regenerates rather than extracts.

My project examines how artificial intelligence can possibly intersect with human and environmental systems to address the maternal mortality crisis in Louisiana, where Black women face disproportionately high risks. Grounded in Black Feminist thought, it explores how AI can be designed to be accountable, life-affirming, and responsive to community needs. Deliverables include a website with AI and data literacy tools and a pilot custom GPT model as a resource for patients and physicians. By connecting maternal health, technology, and social equity, this project investigates how human experience and AI, can come together to sustain and protect vulnerable communities.

A discussion of using Ai as a way to facilitate outdoor, interactive classrooms. Most classrooms within the city of New Orleans already heavily incorporate technology. However, learning is limited to mundane classroom work and excessive discipline. Many of these facilities have restricted outdoor play time rendering students to only 15 minutes in nature, which have negative impacts on classroom behavior and mental health. Additionally, this practice has caused a lack of interest and attention to important subjects, such as science.
My experience as a substitute teacher has shown that many students have trouble expressing complex ideas, especially around sustainability. However, many of them still have valuable insights around the topic.
I propose using Ai as an assistant to help students build out those complex ideas as it pertains to nature and sustainability, while also giving the opportunity to realize their vision with hands on activity within nature. This approach allows them to not only theorize a solution, but offer hands-on learning where they can see their impact in real time.
If nature is a priority then it is imperative that the youth is inextricably entangled into the context of sustainability.

A BUNDL of Ambition & Care is Clay Politz’s story of growth, resilience, and transformation. Through lived experience, he reveals how ambition and care cut through the noise of résumés, degrees, and titles—unlocking the qualities that truly build people and businesses. This journey leads into BUNDL, Clay’s tech company designed to deliver industry-specific AI solutions with humanity at the core. Anchored in his framework—Build it, brand it, market it, sell it, retain it—Clay shows how ambition and care aren’t just values, but practical tools that shape companies, leaders, and the future of work.