search
LLM applications
Trends
- 1Developers urged to run their own LLM proxies for transparency●If you don't run your own LLM proxy, how do you know whats going on?
A new article argues that developers relying on third-party tools to access large language models have little visibility into what data is being sent and returned. The piece makes the case for running your own LLM proxy, letting users log and inspect traffic between their applications and AI providers. Discussion among developers centres on privacy, debugging, and trust in AI tooling.
- 2Free agentic app builder limits LLMs to what tools can't do●A free agentic app builder where LLMs only write what deterministic tools can't
A free browser-based agentic app builder from the Besser-PEARL project is drawing attention for its hybrid approach: large language models generate only the parts of an application that deterministic tools cannot produce, while conventional code generation handles the rest. The pitch has sparked discussion among developers about whether constraining AI to gaps in deterministic tooling could make AI-assisted app building more reliable and reproducible.
- 3Evaluating LLM systems at scale with DeepEval●Beyond chat demos: evaluating LLM systems at scale with DeepEval
A World Bank blog post argues that chatbot demos are not enough to judge large language model systems, and outlines how to evaluate them at scale using the open-source DeepEval framework. It walks through structured, automated testing of LLM outputs for real-world applications, a topic gaining attention as organisations move AI systems from prototypes into production.
Repos
- mlabonne/llm-course Course to get into Large Language Models (LLMs) with roadmaps and Colab notebooks.