Index additions: 2026-W42
Resources added to the index in this ISO week. This is a website archive, not a sent newsletter.
141 resources · page 3 of 5
Prompting
Prompting best practices (Claude docs)
by Anthropic
Current-model guidance in one page: XML structure, examples, thinking, agentic use. (Old use-xml-tags, multishot and extended-thinking-tips URLs redirect here.)
Prompting Fundamentals and How to Apply them Effectively
Practitioner write-up applying core techniques to real tasks.
Structured outputs (Gemini API)
by Google
Gemini's equivalent: constrain responses to a JSON schema.
Prompt engineering how-to (Llama docs)
by Meta
Official guide for prompting open-weight Llama models.
GPT-5 prompting guide
by OpenAI
Model-specific advice on agentic eagerness, tool preambles and instruction conflicts.
Reasoning best practices
by OpenAI
When to pick a reasoning model and how prompting them differs from chat models.
Structured model outputs
by OpenAI
How to get schema-valid JSON instead of hoping the model formats it right.
grok-prompts (published system prompts)
by xAI
Inspect system prompts published by their original provider.
Prompt Engineering
Dense, research-backed survey of prompting methods with citations.
The Prompt Report: A Systematic Survey of Prompt Engineering Techniques
A broad research taxonomy for comparing prompting techniques and finding their sources.
DSPy
by Stanford NLP
Build modular model programs and improve them against a metric.
Self-Consistency Improves Chain of Thought Reasoning in Language Models
Sample several reasoning paths and take the majority answer.
Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
Early research on using worked reasoning examples to improve performance on reasoning tasks.
Tree of Thoughts: Deliberate Problem Solving with Large Language Models
Search over branching reasoning steps; basis for many planning approaches.
Evaluation
AI Evals FAQ
by Hamel Husain and Shreya Shankar
Practical answers about error analysis, annotation and reliable evaluators.
Your AI Product Needs Evals
by Hamel Husain
Build a feedback loop from real failures, not general benchmark scores.
SWE-bench Leaderboards
by SWE-bench team (Princeton/Stanford)
Understand coding-agent scores in terms of resolved repository tasks.
lm-evaluation-harness
by EleutherAI
Run model benchmarks through reusable task and backend interfaces.
Evaluating the Effectiveness of LLM-Evaluators (LLM-as-Judge)
Compare model-based graders against the judgments they must approximate.
simple-evals
by OpenAI
Read reference evaluation code with an explicit maintenance caveat.
Inspect
by UK AI Security Institute and Meridian Labs
Compose evaluation datasets, agents, tools and scorers.
Foundations
The Illustrated Transformer
by Jay Alammar
See how attention and encoder-decoder blocks fit together.
Neural Networks: Zero to Hero
Learn neural networks by implementing them in Python.
Hugging Face LLM Course
by Hugging Face
A structured route through the Hugging Face ecosystem.
LLMs-from-scratch (book code)
Follow a GPT implementation rather than treating the model as a black box.
Applied LLMs: What We've Learned From A Year of Building with LLMs
by Yan, Bischof, Frye, Husain, Liu, Shankar
Cross-functional lessons from building language-model applications.
Lil'Log
Research learning notes that connect methods to their literature.
Model Training
Qwen2.5 LoRA fine-tuning walkthrough (Chinese)
by Datawhale self-llm contributors
Follow a concrete Chinese LoRA recipe from instruction records to a trained adapter and inference.
ruGPT training and fine-tuning examples
Inspect Russian-model fine-tuning and pretraining scripts, with explicit warnings about the legacy environment.
GigaChat Family: Efficient Russian Language Modeling Through Mixture of Experts Architecture
by GigaChat team / SaluteDevices
Study a Russian-origin model team’s account of mixture-of-experts architecture, pretraining and evaluation.
LlamaFactory
by hiyouga (Yaowei Zheng) et al.
Compare adaptation methods within a shared training interface.
TRL - Transformers Reinforcement Learning
by Hugging Face
Find training interfaces for supervised and preference-based adaptation.