Turning complex reality into simple products.

WHY I BUILT THIS

Despite growing mental health awareness, people still struggle. They struggle to talk to one another, and they struggle to seek support. Even those who find support often need help in the moments between sessions, and until now the only thing filling that gap was a crisis hotline. I wanted to make mental health support more accessible: to people who don't feel comfortable sharing their struggles with another person, to those who have support but lack guidance in the hardest moments that happen outside of a session, and to those who can't afford human support.

WHAT IT DOES

Selvenn is an AI life coaching platform for relationships, goals, and everyday life. It started as a conversation-analysis tool for couples and grew into a full coaching platform. Users talk to Lena, the AI coach, through voice sessions and chat, upload files, journal, and complete personalized tasks. Selvenn surfaces insights, explains why those moments happen, and helps users improve, with clear visibility into their progress over the long term. Progress is gamified as a levelled journey that unlocks deeper insights along the way. The vision: Selvenn is an extension of our brain that helps us live mentally and physically better lives. Live with active users since November 2025.

WHY I BUILT THIS

Customer acquisition is difficult, especially for early-stage companies. Paid advertising is fast but expensive, while organic growth and word of mouth are more authentic but take significant time and effort to build. I saw an opportunity to make this middle ground more efficient — helping companies participate in real conversations at scale, without losing authenticity. Platforms like Reddit and Quora already contain high-intent discussions, but finding the right ones and responding thoughtfully is time-consuming. ThreadLift was built to make this process faster, more structured, and easier to scale.

WHAT IT DOES

ThreadLift identifies relevant conversations across platforms like Reddit and Quora and helps you engage in them effectively. For each thread, it generates context-aware reply drafts that you review and refine before posting. The goal is not to automate engagement, but to make high-quality participation more efficient. ThreadLift helps teams reduce the time required to find and engage in relevant discussions, contribute in a way that feels natural to each platform, and build awareness through consistent, credible participation rather than paid campaigns. It is designed for founders, marketers, and teams who want a more scalable approach to organic distribution. Live since April 2026.

WHY I BUILT THIS

There is growing interest in personal health optimization, driven by figures like Bryan Johnson and Andrew Huberman, as well as the rapid adoption of wearable devices. In practice, most tools either overwhelm users with raw data or simplify the problem to the point where the insights are not actionable. What felt consistently missing was a clear connection between everyday behavior and how you actually feel over time. I built LifeMadeEasy to focus on that gap — making it easier to understand how daily actions, routines, and context influence personal health, without requiring users to interpret complex data themselves.

WHAT IT DOES

LifeMadeEasy integrates data from sources such as Apple Health, Oura Ring, and calendar activity to build a unified view of your day. It analyzes this data to identify relationships between behavior and physiological signals — for example, how different types of days or meetings correlate with heart rate and stress, whether specific routines are associated with better sleep or recovery, and which activities tend to align with higher or lower HRV over time. The system focuses on surfacing patterns and correlations in a way that is interpretable and useful, rather than presenting raw metrics alone. An AI assistant allows users to log activities conversationally and explore their own data through questions, while a timeline interface presents daily life as a continuous sequence of events and signals that can be reviewed and annotated.

WHY I BUILT THIS

I wanted to see the systems I build and the knowledge I accumulate as living, spatial structures I can walk through — not flat documents or scattered notes. This is a personal exploration, not a product, but it's live and anyone curious can try it.

WHAT IT DOES

A 3D interactive visualization environment for system architecture, timelines, and knowledge graphs. Map a microservices architecture and watch data flow across layers, plot project history on a spiral timeline synced to Google Calendar, or build a knowledge graph connecting projects, skills, people, and ideas in 3D space. An AI assistant lets you build and modify visualizations through natural conversation. Every space is embeddable publicly with granular admin controls — all configurable via chat. You can also train the AI on your own knowledge and essentially chat with yourself — a personal thinking partner that mirrors how you reason. Next up: letting these "clones" take on real work autonomously. Built with raw Three.js, Next.js, and PostgreSQL. AI features powered by Claude; developed end-to-end with Claude Code. Live at lifeos.solutions.