Notes from the build
Engineering write-ups on AI systems, vector search, automation pipelines and web performance — written while building the projects they describe.
Search by concept
Queries run through an embedding index, so related ideas surface even when the wording differs.
Articles are embedded into a vector index. Your query is embedded the same way, and the closest passages come back ranked by similarity — so “keeping models honest” can surface a post that never uses those words.
Enter a concept to begin
All articles
- 01
Building an AI-Powered Ad Creator with Python & Gemini 2.0
Video content is king, but producing it at scale is expensive and time-consuming. What if you could just type a company name and get a marketing video in minute
PythonGemini 2.0AI VideoFFmpegAutomation2026-02-18→ - 02
Automating Venture Capital Scouting with Llama 3 & Jina AI
Venture Capital relies heavily on data, but finding and analyzing early-stage startups is often a manual, tedious process. I wanted to see if I could automate t
Next.jsLlama 3Jina AIWeb ScrapingGroq2026-02-15→ - 03
ResumeFit: Intelligent Resume Matching with TF-IDF & Scikit-Learn
## The Problem with Black Boxes
DjangoNLPMachine LearningPython2026-02-14→ - 04
Building a Production-Ready Binance Futures Trading Bot in Python
## Introduction
PythonTradingBinanceCLI2026-02-14→ - 05
Automating Content Creation: A 100% Free AI Video Pipeline
## The Vision: One Topic, One Video
n8nAIAutomationPython2026-02-14→ - 06
Mastering WhatsApp Automation with n8n and PostgreSQL
We've all been there—messages pinging at 2 AM, or getting distracted by chats when you're in deep work mode. I decided to solve this by building a **Smart Whats
Automationn8nWhatsApp APIPostgreSQLWorkflow2026-02-12→ - 07
Rendering 50k+ Rows: Optimizing React Data Grids
Handling large datasets in the browser is a classic frontend engineering challenge. When you dump 50,000 DOM nodes into a page, the browser grinds to a halt. In
ReactPerformanceClean CodeFrontend2026-02-08→ - 08
Engineering a Cinematic Portfolio: Framer Motion, RAG, and CI/CD
Building a portfolio is often a tradeoff between showing *what* you built and showing *how* you build. For version 2 of my portfolio, I wanted to bridge that ga
Next.jsFramer MotionGitHub ActionsUI/UXRAG2026-02-06→ - 09
Why I Built RepoSense: Analyzing GitHub Activity
Developers love data. We love optimized algorithms, fast databases, and clean code. But oddly, we often lack data on *how* we work. That's why I built **RepoSen
Open SourceNext.jsAnalyticsGitHub API2026-02-05→ - 10
Implementing Semantic Search in a Next.js Portfolio with Pinecone and Google AI
## The Problem: Beyond Keyword Search
AISemantic SearchPineconeNext.jsVector DatabaseGoogle AI2026-02-03→ - 11
Fraud Detection Using Machine Learning
## Detecting Fraud with Data, Not Guesswork
Machine LearningFraud DetectionPythonData Science2026-01-29→ - 12
Building a Real-Time Multi-User Face Recognition System Using LBPH
## Design and Implementation of a Real-Time Multi-User Face Recognition System Using LBPH
Computer VisionC++OpenCVFace RecognitionAI2026-01-28→ - 13
Lessons Learned While Debugging Real Next.js Production Errors
## Debugging Is Where Real Learning Happens
Next.jsDebuggingWeb DevelopmentProduction2026-01-27→ - 14
Implementing SEO in Next.js: Sitemap, RSS, and Robots.txt
## Why SEO Matters Even for a Portfolio
Next.jsSEOWeb DevelopmentApp Router2026-01-26→ - 15
Building My Portfolio with Next.js App Router
## Why I Decided to Build My Own Portfolio
Next.jsPortfolioFrontendApp Router2026-01-25→ - 16
AI ML and Deep Learning clearing the confusion once and for all
## Understanding AI, Machine Learning, and Deep Learning as a hierarchy rather than competing terms—from the broad AI umbrella to data-driven ML to neural-netwo
AIDeep LearningML2026-01-25→
Start a conversation
Send a note about a role, a build, or something you'd like to take apart together.