Click any tag below to further narrow down your results
Links
The author argues that coding agents, asked to build software that itself calls LLMs, reliably reach for brittle non-LLM approaches like regex or symbolic solvers. He offers three possible explanations and leans toward the idea that current agents are simply poor at software design.
- In one vibe-coded memory-contradiction tool, the agent built a complex symbolic logic system that couldn't work, even though it later admitted the underlying problem was a 130-year-old unsolved one; a single LLM call would have been the right tool.
- The author's own Autodeck agents kept categorizing flashcards with regex and keyword search instead of making another LLM call.
- Three candidate explanations are weighed: training code predates LLMs (ruled out, since models don't default to jQuery-era patterns either), RL environments rarely expose LLM API access, and agents are just bad at software design, which the author now finds most convincing.
This article breaks down the core concepts behind LLMs—from next-token prediction training to tokens, vectors and attention layers—to show how they generate text. It also covers context windows, parameters and why model scale affects performance.
- LLMs work purely through next-token prediction learned by hiding and guessing words billions of times during training, not through actual understanding of letters or math.
- Context windows now range wildly, from 200K tokens (~150K words) up to Llama 4 Scout's 10 million tokens, but bigger windows don't fix factual errors or logical gaps.
- Because models process text as tokens/vectors rather than raw letters, they inherently struggle with tasks like counting letters or doing arithmetic.
The article discusses the importance of data activation in enhancing the performance of large language models (LLMs), particularly in the healthcare sector. It highlights recent advancements in transforming structured medical data into usable formats for LLMs, emphasizing the need for effective reasoning methods to fully leverage the potential of healthcare data.
- Having proprietary data is no longer enough—the real advantage comes from "activating" it into forms LLMs can actually use before competitors catch up
- Tables2Traces converts structured medical data into reasoning traces via contrastive reasoning, notably boosting LLM performance on medical tasks
- Doctors have questioned the fidelity of these synthetic reasoning traces, and gains so far appear mainly in weaker models, raising doubts about scalability
- Despite heavy healthcare-focused LLM investment from OpenAI and Anthropic, the field is fragmented and the best method for transforming healthcare data (knowledge graphs, ontology grounding, etc.) is still unsettled
Non-programming leaders starting to contribute to code with LLMs can increase iteration speed and introduce diverse perspectives, but this also risks compromising the implicit architecture of the codebase. As more non-engineers make changes, maintaining design intent and code maintainability becomes a challenge, requiring developers to adapt their roles to focus on architectural oversight. Despite these risks, democratizing coding could lead to better solutions as more perspectives are included in the development process.
- LLMs are lowering the barrier enough that non-engineer bosses/founders are now directly committing code, not just prototyping
- The real risk isn't code quality but erosion of implicit architecture—decisions and conventions that were never written down and that LLMs can't infer
- This forces a shift in developers' roles toward explicit architectural oversight and codifying design intent so it survives contact with non-engineer contributors
- Despite the risks, wider participation in coding could surface better solutions by incorporating more diverse perspectives into development