.png)
2 days ago
2
Hi Friends,
Happy Friday.
Here are some links (and descriptions) for some things that I found interesting.
Alright, on to the links:
Quick tip for anyone building AI agents - stop making this expensive mistake
Seeing way too many developers burn through API credits because they're not caching responses properly. If your agent is asking the same question multiple times in a conversation (like "what's the user's timezone" or "what are their preferences"), cache that stuff locally instead of hitting the API every single time. Simple Redis cache or even just storing it in memory for the session can cut your API costs by 60-70%…An AI agent is an LLM wrecking its environment in a loop
Solomon Hykes just presented the best definition of an AI agent I've seen yet, on stage at the AI Engineer World's Fair…The Compound Error Crisis: Why LLM Agents Are Failing Like Broken Robots (And Why Computer Science Warned Us)
A 6-axis robot arm reaches for a coffee cup. Joint 1 is off by 0.5 degrees. Joint 2 compensates but overshoots by 0.8 degrees. By the time the arm reaches the cup, it's 3 inches to the left and crashes into the table….An LLM agent analyzes quarterly sales data. It misinterprets Q2 growth as 15% instead of 5%. This becomes the baseline for Q3 projections. The agent then builds a hiring plan based on the inflated projections. By step 5, it's recommending the company triple its workforce….
Both scenarios showcase the same fundamental problem: error propagation. Yet while computer science theory predicted this decades ago and robotics engineers have spent decades developing sophisticated error correction mechanisms, the AI community is deploying multi-step LLM agents with barely a whisper about compound failures…
AI-Native OS (1/3): Principles
This is the first part of a series based on our experiences at MadKudu, where we've shifted to treating AI agents as full-time developers, not just assistants. We found that traditional development practices break down completely under this model. I've distilled our learnings into 11 principles that have made AI-native development practical and significantly boosted our team's productivity (e.g., by focusing on mono-repos, strict typing, and redefining the engineer's role towards documentation and architecture)…AI Agents Truth Nobody Talks About — A Tier-1 Bank Perspective
Over the past 12 months, I’ve built and deployed over 50+ custom AI agents specifically for financial institutions and large-scale tier-1 banks. There’s a lot of hype and misinformation out there, so let’s cut through it and share what truly works in the banking world…
Zapier: We now have more AI agents than employees (Twitter/X Thread)
Interesting milestone at Zapier: We now have more AI agents than employees…We’re now using 800+ AI agents internally, with our team still under 800. Curious how this works? I pulled together my top 5 agents by department 👇…I built the same customer service agent in 12 different agent frameworks (Twitter/X Thread)
Rogerio Chaves, who is building LangWatch, an LLM Ops platform on evaluation capabilities and user-friendly collaboration, tweeted a thread out detailing how they had built the same customer service agent in 12 different agent frameworks and they were going to share their thoughts on the typescript ones…AI Killed The AI Engineer
Louis Knight-Webb, from the AI Tinkerers club in London and a YC alum wrote this article…Following the release of ChatGPT, a group of people emerged calling themselves AI Engineers. They specialised in building applications around LLMs, often discovering performance improving techniques by probing undocumented behaviour. Today's state of the art models perform most of these tricks natively, without the need for external tricks. This realisation struck me when I noticed that my job over the past six months largely involved removing components we had previously developed. Let's delve into how AI Engineers became the first job to be replaced by AI….
.
.
* Based on unique clicks.
** Find last week's issue (2025-04-11) here.
I hope you had a great week. I’ll see you next Friday!
Check out our archives for more AI Agents, AI Engineering, & LLM Systems treats.
All the best,
Sebastian Gutierrez
https://x.com/seb_g
https://sebgnotes.com
