
Problem solving is what gets me out of bed. Not the abstract kind—the concrete, messy kind where you're staring at a broken system and have to figure out why.
I've done this across domains. Warehouse automation at BairesDev. Early diabetic retinopathy detection at Universidad Americana. Now conversational AI at Saywise. The pattern is the same: hard problem, real constraints, think your way through. That's what energizes me. Not the title. The moment when something clicks and the system works.
I used AI to create a timing app for are.na. The brief was narrow from the start: make the timing experience belong in are.na, instead of building a generic timer and adding the name afterward.
What I liked about this project was the scale. It was a specific idea that I could turn into software without pretending it needed to become a platform. AI helped me get the idea into working form, while the decisions about what the app should be stayed with me.

Anthropic draws a clean line between two tools that look similar but solve different problems. MCP connects Claude to data—credentials, live systems, writes that persist. Skills teach Claude what to do with that data—the recipes, not the kitchen. I tested this distinction against the reference server repo and found the clearest case study: sequential-thinking should have been a Skill, not an MCP server. Here's why. Sequential-thinking provides a tool for structured problem-solving through dynamic reasoning—it manages thoughts, revisions, branches, and step counts. But it fetches nothing. No
The most useful thing AI helped me do this month was accelerate debugging on a tricky real-time streaming pipeline. We were building live event processing for video interactions at Saywise, and there was a latency spike we couldn't immediately trace. I fed the logs and architecture into Claude, walked through the reasoning together, and within minutes had three hypotheses I hadn't considered—one of them turned out to be the culprit: a subtle ordering issue in our Kafka consumer group.
What struck me wasn't that Claude solved it alone. It didn't. But it compressed the time between "something's wrong" and "here's what to check" from hours of rubber-ducking to maybe twenty minutes of focused conversation. I also ran some quick exploratory queries with ChatGPT for syntax validation and used Cursor for inline code suggestions as we tested fixes. For someone who's spent a decade building systems, that's the real win—not replacement, but thinking partners that are always available and never tired of the same problem.
At Saywise, I'm currently an AI Software Engineer, focused on building advanced conversational AI agents. My work involves using cutting-edge machine learning and natural language processing technologies. I've been integrating state-of-the-art AI technologies like OpenAI, ElevenLabs, and Deepgram to improve our AI processing capabilities and response quality.
I've also been involved in building live event and streaming processing initiatives, optimizing real-time video interactions to enhance user engagement. A key part of my role is architecting scalable AI infrastructure designed to handle high-volume conversational interactions with low latency.
From 2019 to 2022, I was the Lead Mobile Engineer at Luke's Local in the San Francisco Bay Area. I led the development of their online food ordering application for iOS and Android, focusing on improving both user experience and operational efficiency. We used React Native, Typescript, and Redux to build it.
During my time there, I integrated advanced analytics using Firebase and Mixpanel to track and improve user satisfaction. I also led the development of a successful referral program that encouraged user engagement and growth. To streamline the supply chain, I designed and implemented a companion application for tracking inbound supplies.
I also refactored the application architecture, which significantly improved performance and reduced loading times, enhancing customer satisfaction. Automating the deployment pipeline ensured continuous and seamless delivery of updates, maintaining high standards of quality and reliability.
During my time as a Senior Mobile Engineer at Voice from 2021 to 2022, I focused on building their NFT marketplace application. It was designed to be carbon neutral and user-friendly on the EOS blockchain, serving over 10,000 customers. I worked with Android, Kotlin, MVVM, and OOP to bring this vision to life.
A significant part of my role involved leading the development of the authentication infrastructure. I also led the integration of KYC services, which was crucial in significantly reducing the number of bot accounts on the platform. Additionally, I designed and implemented the NFT multimedia viewer, enhancing the user experience.
Jordan is the kind of engineer every team hopes for. He pairs deep technical skill — especially in AI and machine learning — with genuinely sound judgment, the sort of person you can hand a vague, thorny problem and trust to come back with something elegant and reliable. What sets Jordan apart, though, is how he carries it: calm under pressure, endlessly generous with his knowledge, and always willing to stop and help a teammate get unblocked. He raises the standard for everyone around him without ever making it about himself. I've learned a great deal working with Jordan, and recommend him wholeheartedly.