@Tebza
Tebogo Tseka
@Tebza
Building with AWSomeness
Joined June 2023
About
Nothing here yet.
Available for
Nothing here yet.
Tebogo Tseka's blogs
Recently published
Building an LLM Judge That Doesn't Lie to You
Mar 31 · 9 min read · Our first LLM judge gave a 9/10 to a page where the hero text was completely invisible. Dark grey text on a dark background image. The CSS was syntactically valid. The HTML was well-structured. Every tag was correct. The page was unusable. And our ju...
Join discussion5 Models, 467 Actions, 1 Winner — What We Learned Comparing LLMs on Real Code Generation
Mar 30 · 10 min read · We tested five AI models on the same task 467 times. Each run produced a complete deployable website — not a code snippet, not a function, not a patch. A real site with HTML, CSS, JavaScript, and assets. The question: can cheaper models match Claude ...
MEAamer and 1 more commentedBeyond Text: How We Built an Evaluation Framework for Multi-File AI Outputs
Mar 30 · 9 min read · Most LLM benchmarks evaluate text. HumanEval checks if a function passes unit tests. SWE-bench measures whether a model can patch a repository. MBPP scores single-function completions. None of these work when your AI agent generates an entire website...
Join discussion