Today

Hey, I'm Nikolas. I'm an AI product engineer who's always curious about how we build things – and how AI is making us rethink it all. I believe good products come from understanding the whole picture, from the big idea to the final code.

This is where I share my journey and ideas, hoping to find other folks passionate about building what's next.

Latest

2026

I gave my blog a Markdown twin for agents

Jun 17

I was regenerating Claude's answers just to copy them

Apr 25

LLMs are shortsighted and that's your fault

Apr 15

I built a statusline because /usage is a distraction

Apr 7

The leaked code isn't the story. The response is.

Apr 1

Browse all posts →

Projects

2026

Yapper

Native macOS app for offline, privacy-first speech-to-text transcription.

running

Journey

Idea →

Dev →

Launched →

Revenue →

Exited

Stack

Swift SwiftUI Whisper macOS

Visit

2025

LucidCode

Turn fuzzy app ideas into structured specs AI tools can actually build.

running

Journey

Idea →

Dev →

Launched →

Revenue →

Exited

Stack

TanStack Start PostgreSQL AI SDK OpenAI Bun

Status

Launched and live. Gathering early users.

Visit

Employee Evaluation Tool

Internal company tool for managing employee evaluation processes.

exited

Journey

Idea →

Dev →

Launched →

Revenue →

Exited

Stack

TanStack Start TypeScript PostgreSQL Google Gemini AI SDK Bun

Status

Successfully sold to my current company. Actively used by 100+ employees for internal evaluation processes.

2024

AI Agents List

Curated collection of the best AI agents from across the internet.

running

Journey

Idea →

Dev →

Launched →

Revenue →

Exited

Visit

Browse all projects →

Open Source

Tokenius

A coding agent built from scratch in the terminal — no frameworks, just TypeScript and LLM calls — to understand how tools like Claude Code actually work under the hood.

TypeScript LLM CLI

Bunbase

Strict TypeScript starter optimized for AI-assisted development — five parallel checks on every commit so agents get instant feedback.

Bun TypeScript oxlint Knip Lefthook

Agentic RAG

Production-grade RAG that goes beyond naive retrieval — hybrid search, LLM reranking, query reformulation, and explicit failure handling.

TypeScript Bun Vercel AI SDK PostgreSQL pgvector

MCP Prompt Lab

Prompt evaluation workbench as an MCP server — run evals, compare providers, and test assertions directly from Claude Code.

TypeScript MCP Evals

Tech Radar Agents

Five AI agents collaborate to evaluate technology adoption decisions — no frameworks, just TypeScript, OpenAI SDK, and ~500 lines of code.

TypeScript OpenAI SDK

Browse all explorations →

Connect

If you want to get in touch with me about something or just to say hi, reach out on social media or send me an email.