BLOGS
Writing about what I learn.
Notes on distributed systems, AI engineering, and the messy middle where they meet — pulled live from my Medium.
Aug 20, 2026
10 min read
From Autocomplete to Autonomy: How LLMs Actually Become Agents
The engine is not the car Somewhere in the last two years, the word “agent” quietly took over every AI conversation. Everyone’s building agents. Everyone’s hiring for agent architecture. Every product roadmap has an “age
Read on Medium
GENAI
Aug 18, 2026
9 min read
Spawning Subagents: How Coordination Actually Works Under the Hood
Multi-agent systems sound abstract until you see the actual mechanics — a specific tool name, a specific field in a JSON response, a specific moment where two paths diverge into “parallel” versus “sequential.” This artic
Read on Medium
ARITIFICIAL-INTELLIGENCE
Aug 18, 2026
6 min read
Same Brain, Different Amount of Rope: What Actually Makes an “Agent” Different From the Chat You’re…
Same Brain, Different Amount of Rope: What Actually Makes an “Agent” Different From the Chat You’re Using If you’ve ever used Claude in a browser tab to ask questions, then heard people talk about “AI agents” doing entir
Read on Medium
May 23, 2026
7 min read
What Actually Happens When You Type Into ChatGPT or Claude From Keystroke to Answer?
LLM’s In Action To most users, typing a prompt into Claude or ChatGPT feels like texting a hyper-intelligent human sitting on the other side of the screen. You press enter, a loading bubble blinks for a split second, and
Read on Medium
LARGE-LANGUAGE-MODELS
May 22, 2026
4 min read
The Mechanics of Creativity: How Temperature Hijacks LLM Outputs
To understand how an AI controls its balance between rigid logic and wild creativity, we need to trace the exact journey a piece of data takes. We will follow a single token from the moment it leaves the dark depths of t
Read on Medium
May 22, 2026
5 min read
The Secret LLM Parameter You Aren’t Using (And It’s Not Temperature)
LOGITS IN ACTION Every time you type a prompt into ChatGPT or Claude, a miracle of statistics occurs. The model looks at your text and, word by word, constructs a flawless response. To the end-user, it feels like magic.
Read on Medium
LEETCODE
Apr 25, 2024
2 min read
Dynamic Programming Basics
Optimal substructure: Optimal substructure is a principle mainly used in the context of dynamic programming and some greedy algorithms. It refers to a situation where the optimal solution to a problem can be constructed
Read on Medium
LOW-LEVEL-DESIGN
Apr 13, 2024
2 min read
Interface Seggregation Principle
ISP aims to reduce the side effects and frequency of required changes by splitting a wide set of actions into smaller and more specific ones. The principle encourages designing interfaces that are client-specific rather
Read on Medium
SOLID-PRINCIPLES
Apr 12, 2024
3 min read
Liskov Substitution Principle
Objects in a program should be replaceable with instances of their subtypes without altering the correctness of that program. In simpler terms, if class B is a subtype of class A, then we should be able to replace A with
Read on Medium
Apr 5, 2024
2 min read
Transitioning to Python in LeetCode: A Personal Journey
Starting on my coding journey in 2021, C# served as the language for my inaugural LeetCode problem. Gradually, I moved to Java and started solving LeetCode challenges. However, an enlightening experience shifted my traje
Read on Medium