About

Hi, I'm David, the creator of this app. Thanks for stopping by. I hope you take some time to explore what I've built.

I'm an early-career engineer who created Flow Into Code to solve a problem I ran into myself. If you'd like to connect or follow along with what I'm building, you can find me on LinkedIn and other platforms. I've put a lot of thought and intention into this project, and I'm proud of what it's becoming.

Why I Built Flow Into Code

My inspiration for building Flow Into Code came from failing technical interviews. Not from a lack of preparation or knowledge, but from the time pressure of a real interview. Communication is a critical part of interviewing, yet it's entirely different from how LeetCode problems are usually practiced: alone, quietly, and straight into code.

I built Flow Into Code to practice as close to real interview conditions as possible. The AI interviewer chat aims to replicate the kinds of interactions you'd have with a real interviewer. The structure of each session is designed to encourage a consistent approach to both familiar and new problems, so that consistency carries through when it matters most.

Stripped of the coding disguise, interviews are really about problem-solving ability. They evaluate our ability to translate requirements into solutions, detect edge cases, clarify ambiguities, and create a plan before putting it into motion. That's what Flow Into Code is built around.

A Consistent Process

“Practice how you play.” That saying from sports applies just as well to technical interviews. By creating a consistent flow through the problem-solving process, we're more likely to perform well in an actual interview setting.

Code should be the simplest part of the solution. Data structures and algorithms are developed in a language-agnostic way, so at that point it largely comes down to familiarity with syntax. What matters most is solving the right problem, identifying the right edge cases, and shaping a clear approach before writing implementation details.

Using completed code as a measure of problem-solving ability may be imperfect, but it's the reality. Practicing the translation from approach to code is a key part of succeeding within that system.

Reframing LeetCode Practice

Many people practicing LeetCode are in the early stages of their careers. They may be aiming for internships, new grad roles, or entry-level positions. I understand how frustrating it can be to apply for entry-level jobs that still expect significant experience. At the same time, many industry professionals dislike LeetCode as an evaluation tool because it feels disconnected from their daily work.

Flow Into Code tries to bridge that gap by reframing classic LeetCode problems as real software engineering tasks. Instead of abstract puzzles, problems are presented from multiple perspectives (general, backend, systems) to show how algorithms surface in different industry contexts. The goal is to make the connection between interview preparation and real engineering work feel more concrete.

Why LeetCode Still Matters in 2026

There's a lot of speculation about whether LeetCode or “implementation” skills will still be valued in the age of rapidly advancing AI tooling. My take is that fundamentals will always matter.

The existence of binary search libraries doesn't make it obsolete to understand how binary search works. In a world where more code is written by AI, it becomes even more important to read and understand code, both to learn from it and to review it critically.

AI may use efficient but less common algorithms and techniques. Practicing programming problems exposes us to those patterns so we can recognize and understand them when reviewing AI-generated code.

Even if AI eventually produces most of our code, I would trust someone who understands how to code to prompt and guide an LLM far more than someone unfamiliar with how the output actually works.

Code Execution

Flow Into Code supports basic code execution. The platform does not automatically grade solutions against hidden test cases, so it's up to you to write your own tests in the editor. That mirrors whiteboarding interviews, online assessments with minimal tests, and real development work.

Free. Forever.

I don't plan to charge for this app. Each user gets five daily practice sessions, limited by my monthly LLM and hosting credits. I'll adjust those limits based on what I can sustainably support.

Open Source

This app is open source and available on GitHub. If you'd like to host it yourself to bypass daily limits or make your own changes, you absolutely can.