Skip to content

January 06, 2026 - Daily Drill

🎯 Daily Goals

  • Lumosity training
  • Leetcode
  • Developing laplace
  • Langchain Learning

📝 What I learned:

Interesting Resources:

Summary of AI agent in 2025

AI agents, has become popular since the beginning of the 2025. With the ability to plan, perceive, reason, decide and act, it seems to be the future paradigm of AI systems. It carries the job of transforming the old reactive knowledgable chatbot into something like a digital coworker who can proactively help you solve problems on its own. The prospect is beautiful but we are still not there quite yet. There have been a widespread of AI agent adoption among the companies. Given the stats from claude's report, about 57% of surveyed company incorporated AI agent into their system and 16% of them have progressed to end-to-end processes spanning multiple teams or business functions. While the ubiquitous adoption is profounding, the lag of scaling up can't be ignored. No more than 10% of companies successfully scale up the agent system in any individual function. This suggests that the deployment of agent system is still stuck in the pilot phase. Companies are building agents that works around 80% of the time as a prototype. There is still a long way to go to bridge that gap between 80% to 99.9% reliability. Different industries have different altitude towards the agent system. Suprisingly, the majority of the agent system is not deployed in the technology sector, instead in the financial sector followed by technology and coporate services. This is maybe because works in finance and banking requires comprehending a huge amount of text and structured tubular data which the llm is good at.

Leetcode Problem

Thinking Process

Complexity Analysis

  • Time Complexity: O()
  • Space Complexity: O()

Key Takeaways

🚀 Resources that Requires Further Study

📃 Summary and Reflection