igorfil.com

Blog

Igor Fil's thoughts on technology, programming, and life

  1. Nobody Can Stop Playing

    How AI turned hiring into a self-reinforcing loop: free applications flood recruiters, recruiters reach for keyword filters, candidates game the filters, resumes stop meaning anything, interviews get longer and more …

  2. Pizza Bot - An Open Source Inbox for AI Agents

    Introducing Pizza Bot, the first agentic platform built as an app, not a CLI. No configs needed, everything works with 1 click.

  3. Heard It, Seen It, Done It

    Knowledge comes in three tiers: heard it, seen it, done it. AI is the best compression engine for the first transition and a heavy anchor for the second — it creates fake level 3, rented knowledge. Why renting is the …

  4. The Rock That Bounces Off Water

    Why AI summaries don't stick: effort is the encoding mechanism for knowledge, and when AI removes the friction from research it also removes the learning. The difference between disposable knowledge you can safely …

  5. Me Included

    Why programmers calling AI code bad are the above-average-driver bias in action — good code was always ergonomics for human cognition, and now that agents write and rewrite code faster than humans can read it, the …

  6. Everyone Is Afraid of Missing a Train That Isn't Coming

    Why the Forward Deployed Engineer hiring boom looks like the microservices, Kubernetes, and Kafka hype cycles before it — a solution built for one company's specific problem, adopted everywhere else out of fear of …

  7. Resilience ≠ Slowing Down — Building AI Systems That Survive Contact with Reality

    Why AI startups need Business Impact Analysis and targeted resilience investment. Not every component deserves the same urgency. Learn the five-stage AWS Startup Resiliency Baseline and why security IS resilience.

  8. RAG Is Just a Search Engine Bolted to a Language Model

    Learn how RAG works: embeddings, vector search, hybrid retrieval, and reranking — explained simply. Plus: RAG vs. fine-tuning, and why retrieval beats baking facts into weights.

  9. I Made My iPhone Dumb. Here’s What Changed

    Tired of constant notifications and AI-generated noise? I deleted most apps from my iPhone and lived with a “dumb” phone for two weeks. No FOMO, clearer mind, actual freedom. My raw experience.

  10. Why Startups Need Resilience Early — Introducing the AWS Startup Resiliency Baseline

    Startups often postpone resilience planning until it's too late. The AWS Startup Resiliency Baseline provides a practical framework for building fault-tolerant systems from the start without slowing down your velocity.

  11. AI Startup Resiliency: Five Essential Lessons from Amazon Bedrock

    Real-world lessons from guiding startups in building production-grade AI applications with Amazon Bedrock. These five areas form a blueprint for AI startup resiliency—from authentication reliability to geographic …

  12. The Electron Paradox: Why The Best Desktop Apps Are Just Websites in Disguise

    Electron's inefficiency isn't a flaw—it's the rational choice in a fragmented desktop world. Discover why Microsoft, Discord, and Slack chose web tech over native, and how economic reality beats architectural purity when …

  13. Unit Testing for AWS Applications: Prevent Downtime and Scale with Confidence

    Unit testing is essential for building resilient AWS applications. Learn how automated testing prevents downtime, catches bugs early, and enables startups to scale Lambda, EC2, and RDS confidently without breaking …

  14. Infrastructure as Code for Startups: AWS CloudFormation & Terraform Best Practices

    Infrastructure as Code (IaC) helps startups build resilient AWS systems that scale fast and stay reliable. Learn best practices for CloudFormation and Terraform to create cost-effective, antifragile cloud architectures.