This is part 2 of a research series documenting an attempt to build something adjacent to — and in...
This is part 2 of a research series documenting an attempt to build something adjacent to — and in some ways alternative to — large language models. Part 1: "My Synthetic Eval Said 30/30. LoCoMo Said 0.13." The code that survived lives in AuraSDK.
Not a chatbot wrapper.
Not another prompt stack.
Not a vector database with a nicer UI.
The question was narrower and more dangerous:
10 adversarial scenarios, 64 assertions, 3-tier evaluation pyramid. Llama, Qwen, GPT-OSS — none scored above 63%. Here's what…
How a portfolio-monitoring bot taught me how LLMs really work: zero-temperature non-determinism, determinism in the data, and…
Every production LLM app eats false-positive refusals. A user asks something perfectly fine, the...
I recently shipped a small research agent that beats three recurring LLM problems: stale training data, hallucinated facts, and…
If you have deployed a large language model in production, even just as a personal coding assistant,...
Hello, I'm Shrijith Venkatramana. I'm building git-lrc, an AI code reviewer that runs on every...
