https://blog.wiseprobe.io/Blog and portfolio focused on artificial intelligence, generative ai, machine learning, natural language processing (NLP), large language models (or LLMs), and other technical things. 2026-02-26T13:01:03+00:00 WiseProbe https://blog.wiseprobe.io/ Jekyll © 2026 WiseProbe /assets/img/favicons/favicon.ico /assets/img/favicons/favicon-96x96.png 2026-02-21T00:00:00+00:00 2026-02-26T13:00:37+00:00 https://blog.wiseprobe.io/posts/coding-agent-sdks/ WiseProbe

“Simplicity is prerequisite for reliability.” - Edsger Dijkstra There are dozens of different agentic coding assistants around right now, from Claude Code (Anthropic) and Codex (OpenAI) to Kiro (Amazon) and Goose (Block). Other coding agents include Windsurf, Cursor, OpenCode, and the Pi agent by Mario Zechner. Most use coding agents through a terminal, web, or desktop UI, but having a progr...

2026-02-07T00:00:00+00:00 2026-02-09T01:08:38+00:00 https://blog.wiseprobe.io/posts/building-patchpal/ WiseProbe

LLM coding agents may look complex, but fundamentally they’re just a loop, some tools, and an LLM doing the real work. Agentic coding assistants are now widespread (e.g., Aider, Claude Code, Codex, OpenCode). After spending some time with Claude Code, I wanted something I could run with both local and cloud models, understand end-to-end, and modify/extend without wading through a large framewo...

2024-11-13T00:00:00+00:00 2024-11-13T20:02:18+00:00 https://blog.wiseprobe.io/posts/resume_parsing/ WiseProbe

Resume-parsing is a great use case for generative AI. In this post, we demonstrate how local LLMs like Meta’s Llama-3.1 can extract useful information from resumes to make them more findable. We will use the OnPrem.LLM package, a handy Python toolkit for working with LLMs. By default OnPrem.LLM uses the Llama-cpp -python under the hood. However, OnPrem.LLM can be used with a number of local LL...

2024-10-28T00:00:00+00:00 2024-11-13T20:02:18+00:00 https://blog.wiseprobe.io/posts/onprem/ WiseProbe

Not long after the release of OpenAI’s ChatGPT, data security vendor Cyberhaven observed that employees across industries were sharing sensitive and privacy-protected information while interacting with the large language model. Within the defense industry, usage of such external services poses a significant security risk. However, since the release of open large language models from organizati...