yoeven

More often than not we pre-process data using an LLM before inserting them into a column/row on the database, for example translating text and storing the different languages or running OCR on a document and storing the…

LLMs have changed the way we code, build, and launch a product. Many of these cases are human-in-the-loop tasks like vibe coding or workflows that have a larger margin of error that is acceptable. However, LLMs aren't…

Would devs want a single API that automatically routes to the best LLM for their goal while having basic tools built in like web scraping, web search, etc? Or would devs want way more control like a Lang chain/AI SDK,…

I love using markdown for basically anything static, from blogs to static pages on my site, to some docs and even email templates. I tried using notion as a CMS of sorts for my blogs but it tends to cause more problems…

I missed the good old days when gpt-3.5 was the best model in the market. You had one library, one "framework" and one standard. Now we have awesome models like R1 that rival o1 being 100x cheaper, gemini flash 8b with…

Most TTS models today have a fixed set of voices that are tied to a fixed set of languages. We built a trained TTS model with the concept of accents like we have in real life. You can have a Chinese person speaking…

Nowadays, a common AI tech stack has hundreds of different prompts running across different LLMs. Three key problems: - Choices, picking from 100s of LLMs the best LLM for that 1 prompt is gonna be challenging, you're…

Hey HN! We’ve outperformed the fastest AI company, Groq, in Speech to Text while having a lower WER score and being more feature-rich. Check out the benchmarks and repo!