Notes on MDT - Personal site

The Only Guide You'll Need for GitHub Spec Kit

Mon, 05 Oct 2026 22:47:00 +0330

Not watched yet. Saved to watch later.

Video from Den Delimarsky.

Building Effective Agents

Mon, 05 Oct 2026 22:24:00 +0330

Not read yet. Saved to read later.

Post by Anthropic.

MCP at the LHC: Coordinating the World's Largest Machine

Mon, 05 Oct 2026 22:22:00 +0330

Not watched yet. Saved to watch later.

Video from Georg Šumailov.

Building AI Agents on the JVM

Mon, 05 Oct 2026 22:20:00 +0330

Not watched yet. Saved to watch later.

Video from Anton Arhipov.

Why Agentic Systems Need Ontologies

Mon, 05 Oct 2026 22:16:00 +0330

Not watched yet. Saved to watch later.

Video from Frank Coyle.

The MCP Abstraction Tax

Mon, 05 Oct 2026 22:14:00 +0330

Not read yet. Saved to read later.

Post by Justin Poehnelt.

OpenClaw

Mon, 05 Oct 2026 22:08:00 +0330

Not read yet. Saved to read later.

From the Thoughtworks Technology Radar.

Hermes Agent Full Course: Build and Sell (2026)

Mon, 05 Oct 2026 00:43:00 +0330

Not watched yet. Saved to watch later.

Video from Samin Yasar.

Code Connect 38 (After Coding — Masoud Beigi)

Mon, 05 Oct 2026 00:36:00 +0330

Not watched yet. Saved to watch later.

Video from Code Connect.

What a Paradise? Understanding Agents

Mon, 05 Oct 2026 00:34:00 +0330

Not watched yet. Saved to watch later.

Video from Jadi Mirmirani.

Everything About AI Agents

Sun, 04 Oct 2026 22:26:00 +0330

Video from Maryam Sadeghi.

Takeaways

AI agent elements: LLM, tools, goal, loop, memory.
Types of memory
- Working memory: the current conversation, like RAM
- Procedural memory: how it should behave. Rules and instructions (skills)
- Semantic memory: stable facts we want the agent to always know
- Episodic memory: a timestamped history of past events and conversations

These memories are stored in a database.

The cost gets high. Distill: summarize and pull out a few important things, then store them in the database.

Liveness is not health

Fri, 11 Sep 2026 17:20:00 +0330

The worst Kubernetes outage I keep seeing is not a node dying. It is a slow database, a liveness probe that talks to that database, and a cluster that then kills every pod that was still doing useful work.

Liveness does not mean “the app is healthy.” It means “this process is so stuck that Kubernetes should shoot it.” If you point that gun at a dependency, you will shoot the fleet.

Spring @Transactional does not mean what you think

Fri, 11 Sep 2026 17:10:00 +0330

In review I still see @Transactional used as a blessing: put it on a method, and the database will behave. It will not. Spring starts a transaction when a proxy intercepts an incoming call. If the call never hits the proxy, you have no transaction, no rollback, and a very confusing production bug.

What I believed → what I do now

I used to sprinkle @Transactional on every service method “to be safe.” I now put it on the one method that is the use-case boundary, I write a test that proves rollback, and I treat same-class private calls as a smell.

Java, Spring, Kubernetes

Fri, 11 Sep 2026 17:00:00 +0330

A map of how I actually run backend software: Java on the JVM, Spring as the application layer, Kubernetes as the place it lives. Start here, then open the note that matches the problem in front of you.

This is not a tutorial index. It is the order I would walk a teammate through after an incident.

If you are new here

Spring @Transactional does not mean what you think — where transactions actually start, and why a method calling another method on this silently does nothing.
Liveness is not health — the Kubernetes outage that looks like a crash loop but started as a slow dependency.
Reliability lessons from SQLite — why a tool that solves more problems than it creates stays in production.

The Java course is the longer path if you want to write this stack with me, not only read about it.

Reliability Lessons From SQLite

Fri, 28 Aug 2026 10:56:00 +0330

Talk from SSW 2026.

Takeaways

SQLite is used so much because it solves more problems than it creates.

Notes

DO-178B Software considerations in Airbone Systems and Equipment Certification

it leads to create Fossil: Better Project Management MC/DC

Example: video note

Fri, 28 Aug 2026 10:00:00 +0330

This is a starter video note. Copy it, or delete it once you add your own.

Use note_type = "video" and put the YouTube URL in source so the video embeds at the top. Headings like 12:40 make timestamped notes easy to scan.

Takeaways

The browser event loop is a queue: JavaScript runs one thing at a time, then picks the next callback.
Rendering and I/O are not the same queue as your JS callbacks.
setTimeout(fn, 0) does not run immediately; it waits until the current stack is clear.

Notes

01:15 — Call stack

JS is single-threaded. Each function call is pushed onto the stack and popped when it returns.

Example: book note

Thu, 27 Aug 2026 10:00:00 +0330

This is a starter book note. Copy it, or delete it once you add your own.

Use note_type = "book", creator for the author, and one heading per chapter. Set status to in-progress while you are still reading.

Takeaways

Reliability, scalability, and maintainability are the three reasons we bother with data systems.
There is no single storage engine that is best for every access pattern.
Replication and partitioning are separate problems; most systems need both.

Chapter 1 — Reliable, Scalable, and Maintainable Applications

Reliability means the system keeps working when things go wrong, not only in the happy path.