Kimi-K3
Kimi K3 is an open-weight, 2.8T-parameter native multimodal agentic model with a 1M-token context window, designed for frontier coding, knowledge work, and reasoning tasks.
它针对一个高度专业化、融合结构生物学、深度学习和实验生物化学的新兴角色,提供了结构化的主题分类和六周学习计划。
A comprehensive, community-curated collection of 185+ interview questions and answers 💡 for Protein Origami Designer roles — professionals who computationally design novel protein structures and assemblies from scratch (de novo design) 🧬 or by engineering existing folds, sitting at the intersection of structural biology 🔬, biophysics ⚛️, deep learning-based structure prediction/generation 🤖, and experimental protein biochemistry 🧪.
Protein Origami Designers 🧬 use computational tools 💻 — physics-based design software (e.g., Rosetta-family methods), deep learning structure prediction (AlphaFold-class), and increasingly generative deep learning design methods (diffusion-based and other generative protein design models) 🎨 — to design novel protein sequences that fold into intended, often entirely novel three-dimensional structures and assemblies, for applications spanning enzyme design ⚙️, binder/therapeutic design 💊, biomaterials 🕸️, and nanostructure engineering 📐. The discipline requires deep fluency in protein biophysics fundamentals alongside rapidly-evolving computational design methodology and tight integration with experimental validation 🧪.
This repository covers:
Estimated preparation time: ⏳ 30–50 hours Interview duration: ⏱️ Typically 4–6 rounds (3–5 hours total), often including a structural design whiteboard round 📝 and a computational methods deep-dive round 🧠
Awesome-Protein-Origami-Designer-Interview-QA/
├── README.md
├── CONTRIBUTING.md
├── LICENSE
├── topics/
│ ├── 01-Protein-Structure-Folding-Fundamentals.md
│ ├── 02-De-Novo-Design-Physics-Based-Methods.md
│ ├── 03-Deep-Learning-Structure-Prediction-Generation.md
│ ├── 04-Protein-Protein-Interface-Binder-Design.md
│ ├── 05-Enzyme-
Protein-Origami-Designer-Interview-QA
根据分类、Topic 和编程语言匹配的相似项目。
Kimi K3 is an open-weight, 2.8T-parameter native multimodal agentic model with a 1M-token context window, designed for frontier coding, knowledge work, and reasoning tasks.
Harness Engineering is a methodology for improving coding agent outputs by carefully crafting the environment around them—providing curated context, tools, and executable constraints that encode an organization’s nonfunctional requirements and cumulative lessons.
A 28.9 million parameter language model runs on an $8 ESP32-S3 microcontroller entirely on-device, generating simple stories at about 9.5 tokens per second.