Considerations on deploying LLM-based workflows
2025-02-25
Let me start with a disclaimer. These thoughts come from a startup perspective where funds and personnel are a bit scarce. There’s a lot of generalizations and assumptions made here as well; take them with a pinch of salt. In just a short span of time, the speed of improvements to LLMs, especially the mainstream ones, are short of astonishing. The massive ones are getting smarter, faster, and more accurate. And even better, the open ones, such as Llama, DeepSeek, Gemma, Qwen, etc. are also catching up, which is a good thing, as I’m more interested in them. And for enterprises who are looking into integrating LLMs into their internal workflows, or even products, the options available now are so many it’s quite confusing where to even start. I hope this blog will shed some light on some of these confusions.
AI · Deployment · Genai · Llm · Programming · Software · Systems · Tech
6 minutes
2024-08-25
TL;DR: To my fellow system builders, don’t dismiss it but understand how it works. As the saying goes, “a tool is only as good as the hands that wield it”. As a software craftsman, your tools are important. And equally so, are your skills in using them effectively. ~~ There’s no escaping GenAI nowadays, is there? I’m sure you’ve seen the full spectrum of its effects by now; from total naysayers to skeptics, to cautious optimists, to proponents and fanatics, to full-blown doom-bringers. In the cloud space, the big three cloud providers are “all in” on AI, as you can see in their headlines. Furthermore, there are thousands of AI-powered startups cropping up left, right, and center, with massive venture capital and valuations. If you’re not in the thick of these things, it gives the feeling that if you don’t invest, or apply AI in your business, or products, it’s just a matter of time when you’ll be left behind and eventually relegated into the pages of failures in history. It’s a scary thought. The FUD is real.
5 minutes