nlp

LangGraph 워크플로우 템플릿 (v38) Python 개발자를 위한 LangGraph 기반 AI 에이전트 워크플로우 템플릿 LangChain과 LangGraph를 사용한 Python 기반 AI 에이전트 개발을 위한 실전 가이드입니다. 이 템플릿은 실제 개발자들이 겪는 문제를 해결하기 위해 설계되었습니다. 1. LangGraph 아키텍처 개요 LangGraph는 상태 기반 워크플로우 시스템으로, 다음과 같은 핵심 구성 요소로 구성됩니다: 핵심 구성 요소: Nodes : 워크플로우의 단계 Edges : 노드 간의 연결 State : 워크플로우의 상태 관리 Checkpointing : 상태 저장 및 복원 from langgraph.graph import StateGraph , END from typing …

Traditional search works on keywords. You type "cheap hotel", it looks for documents containing those exact words. Someone asks "affordable accommodation near the beach". Your documents say "budget-friendly lodging by the coast". Zero keyword overlap. Zero results. Search fails. Embeddings fix this. They convert text into vectors of numbers where similar meanings end up geometrically close. "Chea…

What a recent study on ChatGPT, Python, R, and Stata tells us about AI-assisted coding for causal inference The post Can AI write your code? appeared first on Towards Data Science .

How did semantic search evolve from simple keyword matching into modern transformer-based language understanding? This hands-on article builds four generations of semantic search systems step by step using Python. The post From TF-IDF to Transformers: Implementing Four Generations of Semantic Search appeared first on Towards Data Science .

This blog walks through how AI LinkedIn marketing tools are actually being used on the ground, not just what they claim to do. It starts with the basics, then slowly moves into what’s changed over the last couple of years. Outreach is noisier now, responses are harder to get… so the bar is different. There’s […]

▸ Understand-Anything-Public is a new open-source GitHub project that converts your entire codebase into an interactive knowledge graph. If you've ever inherited a 200K-line project with zero documentation, you know the pain — this tool draws you a map instead of forcing you to read file by file. ▸ The same project highlights a growing trend in AI-assisted development: when bugs surface, devs now…

Scientific Data, Published online: 25 May 2026; doi:10.1038/s41597-026-07397-8 A dataset of human and AI-generated rubric evaluations for formative programming assessment

With the rapid evolution of technology, businesses face mounting pressure to ensure their knowledge resources are both accurate and accessible. Capgemini’s AI Agent for Knowledge Managers is at the forefront of this transformation, revolutionizing how companies build and manage knowledge in a smarter, more agile way. The post Capgemini’s AI Agent for Knowledge Managers is shaping the future of en…

Last year, I met a young translator reinventing herself. She studied Translation for five years at a local university. We met online by commenting and engaging on each other's LinkedIn posts. When we met, she was looking for another way to make money. Her agency cut down her rates and couldn't find her as many gigs as before. Who to blame? AI! AI isn't disappointing at translation. But it doesn't…

resLens uses genomic language models to detect antibiotic resistance genes from DNA sequences, performing especially well on long-read data and on some ARGs poorly represented in reference databases.

This is a submission for the Gemma 4 Challenge: Build with Gemma 4 What I Built StudyGemma — an AI-powered study buddy that transforms messy CS notes into polished study material. I built this because I noticed how frustrating it is to have scattered lecture notes, code snippets, and textbook excerpts scattered everywhere with no good way to organize them. StudyGemma solves that with a single int…

I see a lot of claims about which model is "best." Best at what? For whom? At what cost? I got tired of guessing. So I ran my own comparison. The setup I took 500 real queries from my production logs – a mix of: Code generation (120 queries) Document summarization (150 queries) Question answering (180 queries) Creative writing (50 queries) I ran each query through four models using the same promp…

This is a submission for the Gemma 4 Challenge: Write About Gemma 4 Why I Started This Millions of people want to learn English but cannot afford a private tutor. For many families, this is too expensive. So I decided to build one using AI. Not just a simple chatbot but a real English coach for everyone, from a 7-year-old child 👧 to a 60-year-old adult👴. I called it SpeakUp and it is powered by G…

International audience

International audience

Forthcoming/in press

International audience

This is a submission for the Gemma 4 Challenge: Build with Gemma 4 TL;DR: Vitreus is a spreadsheet agent that lets you ask natural-language questions of CSV and XLSX workbooks, then uses Gemma 4 to return an auditable JSON action manifest. The manifest can highlight rows, write values, and add formulas; Vitreus then applies those changes to a CSV snapshot or a real .xlsx file with colors and form…

Every developer using AI tools knows this pain. You open ChatGPT or Claude, excited to solve a problem, and the first 5 messages are just... explaining your project. "It's a Next.js app with MongoDB..." "The folder structure is like this..." "We're using TypeScript and..." By the time the AI actually understands your codebase, you've lost your focus. I got frustrated enough to build something abo…