Many people think building AI agents is mostly about the LLM. In reality, most of the complexity sits in the infrastructure around it. In my previous article(https://aiforproduction.blog/2026/03/12/what-openclaws-architecture-taught-me-about-building-real-genai-systemspart-1/), I explored the architectural challenges behind building systems like OpenClaw — things like always-on infrastructure, memory management, tool execution, and concurrency. Today, I want to share what I … Continue reading What OpenClaw’s Architecture Taught Me About Building Real GenAI Systems(part 2)- The Solution
Tag: llm
What OpenClaw’s Architecture Taught Me About Building Real GenAI Systems(part 1)
Over the past few days, I’ve been exploring OpenClaw and trying to run parts of it locally. As someone working on production GenAI systems, I’m always curious about how different frameworks approach building AI agents that operate in real environments. What stood out while studying OpenClaw is something I keep seeing across many GenAI systems: … Continue reading What OpenClaw’s Architecture Taught Me About Building Real GenAI Systems(part 1)
Will Jobs Exist as we know?(The Fear)
This series started because of a 1–1 conversation with a reportee. One of my reportees asked me a simple question: “With everything happening in AI… should I be worried, and what will happen to jobs in 5-10 years?” It wasn’t a dramatic question. It wasn’t about headlines. It was about career, growth, and uncertainty. That … Continue reading Will Jobs Exist as we know?(The Fear)
How to Think About AI Systems in 2026
Why GenAI, Predictive Models, and Agents Need to Work Together I’ve been working in AI since well before the GenAI wave, and one thing that has become almost annoying in the last couple of years is how quickly every problem gets thrown at an LLM—with the expectation that it will magically solve everything. It won’t. … Continue reading How to Think About AI Systems in 2026
Why Do Multi-Agent LLM Systems Fail? Insights from Recent Research
As GEN AI evolves, multi-agent systems (MASs) are gaining traction, yet many remain at the PoC stage. So this is where research papers and surveys help. Over the weekend, I explored their failure points and found a fascinating study worth sharing. MASs promise enhanced collaboration and problem-solving, but ensuring consistent performance gains over single-agent frameworks … Continue reading Why Do Multi-Agent LLM Systems Fail? Insights from Recent Research
Optimizing RAG Systems: A Deep Dive into Chunking Strategies.
When it comes to today’s AI systems, especially in Generative AI, the main challenge isn’t just building a basic system with, say, 70% accuracy—it’s about pushing that system to over 90% and making it reliable for real-world production. Optimizing RAG (Retrieval-Augmented Generation) systems is essential for reaching this level. Effective chunking, one of the foundational … Continue reading Optimizing RAG Systems: A Deep Dive into Chunking Strategies.
RAG Components – 10,000 ft Level
When you look at GEN AI and specifically LLM's from a usage point of view, we have a few techniques to interact with LLM's. This depends on whether you need to interact with external data or just use LLMs for your different tasks. RAG is an important technique and widely adopted because it helps you … Continue reading RAG Components – 10,000 ft Level
AI For Production 