# code-completion

Code completion is a code-editor feature that suggests or fills in code as developers type.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## Introducing Mellum2: A 12B Mixture-of-Experts Model by JetBrains

DevFeed: [Introducing Mellum2: A 12B Mixture-of-Experts Model by JetBrains](<https://devfeed.tech/articles/introducing-mellum2-a-12b-mixture-of-experts-model-by-jetbrains-7013.md>)

Original publisher: [Read original article](<https://huggingface.co/blog/JetBrains/mellum2-launch>)

Author: Nikita Pavlichenko

Published: 2026-06-01T15:45:17Z

Content type: article

Language: en

Sources: [Hugging Face - Blog](<https://devfeed.tech/sources/hugging-face-blog.md>)

Topics: [Mixture of Experts (MoE)](<https://devfeed.tech/topics/mixture-of-experts-moe.md>), [Low-Latency Inference](<https://devfeed.tech/topics/low-latency-inference.md>), [Inference](<https://devfeed.tech/topics/inference.md>), [Retrieval Augmented Generation (RAG)](<https://devfeed.tech/topics/retrieval-augmented-generation-rag.md>), [Routing (disambiguation)](<https://devfeed.tech/topics/routing.md>), [Benchmark](<https://devfeed.tech/topics/benchmark.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>), [hugging face](<https://devfeed.tech/topics/hugging-face.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [code](<https://devfeed.tech/tags/code.md>), [code-generation](<https://devfeed.tech/tags/code-generation.md>), [coding](<https://devfeed.tech/tags/coding.md>), [inference](<https://devfeed.tech/tags/inference.md>), [latency](<https://devfeed.tech/tags/latency.md>), [low-latency-inference](<https://devfeed.tech/tags/low-latency-inference.md>), [mixture-of-experts](<https://devfeed.tech/tags/mixture-of-experts.md>), [model](<https://devfeed.tech/tags/model.md>), [rag](<https://devfeed.tech/tags/rag.md>), [routing](<https://devfeed.tech/tags/routing.md>), [software](<https://devfeed.tech/tags/software.md>)

### AI overview

JetBrains introduces Mellum2, a 12B-parameter Mixture-of-Experts model trained from scratch on natural language and code. It activates 2.5B parameters per token to provide efficient, low-latency inference for routing, retrieval-augmented generation, summarization, sub-agent tasks, and high-throughput coding features. The model is released under the Apache 2.0 license and is reported to deliver competitive benchmark performance with more than twice the inference speed of similarly sized models.

### Source excerpt

- Mellum2 is a 12B-parameter Mixture-of-Experts model trained from scratch on natural language and code. - The model activates only 2.5B parameters per token, making it efficient for high-throughput, low-latency inference. Mellum2 is can be used for routing, RAG, summarization, sub-agents, high-throughput coding features, and private deployments. - It is released under the Apache 2.0 license.

## Codestral Embed

DevFeed: [Codestral Embed](<https://devfeed.tech/articles/codestral-embed-6986.md>)

Original publisher: [Read original article](<https://mistral.ai/news/codestral-embed/>)

Published: 2025-05-28T12:00:00Z

Content type: article

Language: en

Sources: [Mistral AI Blog](<https://devfeed.tech/sources/mistral-ai-blog.md>)

Topics: [Embeddings](<https://devfeed.tech/topics/embeddings.md>), [Retrieval-Augmented Generation](<https://devfeed.tech/topics/retrieval-augmented-generation.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>), [AI search](<https://devfeed.tech/topics/ai-search.md>), [coding](<https://devfeed.tech/topics/coding.md>)

Tags: [code-completion](<https://devfeed.tech/tags/code-completion.md>), [code-search](<https://devfeed.tech/tags/code-search.md>), [coding](<https://devfeed.tech/tags/coding.md>), [cost](<https://devfeed.tech/tags/cost.md>), [developer](<https://devfeed.tech/tags/developer.md>), [embeddings](<https://devfeed.tech/tags/embeddings.md>), [performance](<https://devfeed.tech/tags/performance.md>), [retrieval-augmented-generation](<https://devfeed.tech/tags/retrieval-augmented-generation.md>)

### AI overview

Mistral releases Codestral Embed, an embedding model specialized for code retrieval. The article presents benchmark results, configurable embedding dimensions and precisions, and trade-offs between retrieval quality and storage cost. It highlights applications including retrieval-augmented generation, semantic code search, similarity search, duplicate detection, clustering, and code analytics.

### Source excerpt

The most powerful AI platform for enterprises. Customize, fine-tune, and deploy AI assistants, autonomous agents, and multimodal AI with open models.

## Meet Cursor Tab and Never Look Back: The Future of AI-Assisted Coding

DevFeed: [Meet Cursor Tab and Never Look Back: The Future of AI-Assisted Coding](<https://devfeed.tech/articles/meet-cursor-tab-and-never-look-back-the-future-of-ai-assisted-coding-5809.md>)

Original publisher: [Read original article](<https://neon.com/blog/tab-coding-cursor>)

Author: Carlota Soto

Published: 2025-04-04T00:13:42Z

Content type: article

Language: en

Sources: [Blog -- Neon Docs](<https://devfeed.tech/sources/blog-neon-docs.md>)

Topics: [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [cursor](<https://devfeed.tech/topics/cursor.md>), [Vibe coding](<https://devfeed.tech/topics/vibe-coding.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>), [Code](<https://devfeed.tech/topics/code.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-assisted-coding](<https://devfeed.tech/tags/ai-assisted-coding.md>), [autocomplete](<https://devfeed.tech/tags/autocomplete.md>), [code-completion](<https://devfeed.tech/tags/code-completion.md>), [cursor](<https://devfeed.tech/tags/cursor.md>), [developer](<https://devfeed.tech/tags/developer.md>), [learning](<https://devfeed.tech/tags/learning.md>), [product](<https://devfeed.tech/tags/product.md>), [side-projects](<https://devfeed.tech/tags/side-projects.md>), [vibe-coding](<https://devfeed.tech/tags/vibe-coding.md>), [vs-code](<https://devfeed.tech/tags/vs-code.md>)

### AI overview

The article introduces Cursor Tab, an AI-assisted coding feature designed to keep developers in control while writing code. It contrasts "tab coding" with more autonomous vibe coding and describes predictive coding and zero-entropy edits, where Cursor anticipates larger code changes and handles predictable actions after the developer's intent is clear.

### Source excerpt

It's crazy to think that "vibe coding" started just weeks ago, with this tweet from Andrej Karpathy: Now, you can get a job as a vibe coder, and Stanford professors are giving lectures on it. That escalated quickly. In its short life, vibe coding has come to mean all kinds of AI-...

## BigCodeBench: The Next Generation of HumanEval

DevFeed: [BigCodeBench: The Next Generation of HumanEval](<https://devfeed.tech/articles/bigcodebench-the-next-generation-of-humaneval-7314.md>)

Original publisher: [Read original article](<https://huggingface.co/blog/leaderboard-bigcodebench>)

Author: Terry Yue Zhuo; Jiawei Liu; Qian Liu; Binyuan Hui; Niklas Muennighoff; Daniel Fried; Harm de Vries; Leandro von Werra; Clémentine Fourrier

Published: 2024-06-18T00:00:00Z

Content type: article

Language: en

Sources: [Hugging Face - Blog](<https://devfeed.tech/sources/hugging-face-blog.md>)

Topics: [LLMs](<https://devfeed.tech/topics/llms.md>), [Benchmark](<https://devfeed.tech/topics/benchmark.md>), [Code](<https://devfeed.tech/topics/code.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>), [test-coverage](<https://devfeed.tech/topics/test-coverage.md>)

Tags: [benchmark](<https://devfeed.tech/tags/benchmark.md>), [code](<https://devfeed.tech/tags/code.md>), [code-completion](<https://devfeed.tech/tags/code-completion.md>), [collaboration](<https://devfeed.tech/tags/collaboration.md>), [community](<https://devfeed.tech/tags/community.md>), [complexity](<https://devfeed.tech/tags/complexity.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [leaderboard](<https://devfeed.tech/tags/leaderboard.md>), [llms](<https://devfeed.tech/tags/llms.md>), [performance](<https://devfeed.tech/tags/performance.md>), [research](<https://devfeed.tech/tags/research.md>)

### AI overview

The article announces BigCodeBench, a benchmark for evaluating the programming capabilities of large language models on practical, challenging tasks. It contains 1,140 function-level tasks using 139 libraries, with detailed instructions, interactive examples, runtime test harnesses, and extensive branch coverage. The benchmark measures whether models correctly generate code and compares its complexity and tool-use diversity with existing programming benchmarks.

### Source excerpt

We're on a journey to advance and democratize artificial intelligence through open source and open science.

## Codestral

DevFeed: [Codestral](<https://devfeed.tech/articles/codestral-6991.md>)

Original publisher: [Read original article](<https://mistral.ai/news/codestral/>)

Published: 2024-05-29T09:00:00Z

Content type: article

Language: en

Sources: [Mistral AI Blog](<https://devfeed.tech/sources/mistral-ai-blog.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Generative AI](<https://devfeed.tech/topics/generative-ai.md>), [Code generation](<https://devfeed.tech/topics/code-generation.md>), [coding](<https://devfeed.tech/topics/coding.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>), [Benchmark](<https://devfeed.tech/topics/benchmark.md>), [context window](<https://devfeed.tech/topics/context-window.md>), [Python](<https://devfeed.tech/topics/python.md>), [Java](<https://devfeed.tech/topics/java.md>), [C](<https://devfeed.tech/topics/c.md>), [C++](<https://devfeed.tech/topics/c-plus-plus.md>), [JavaScript](<https://devfeed.tech/topics/javascript.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [bash](<https://devfeed.tech/tags/bash.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [c](<https://devfeed.tech/tags/c.md>), [c-plus-plus](<https://devfeed.tech/tags/c-plus-plus.md>), [code](<https://devfeed.tech/tags/code.md>), [code-completion](<https://devfeed.tech/tags/code-completion.md>), [code-generation](<https://devfeed.tech/tags/code-generation.md>), [coding](<https://devfeed.tech/tags/coding.md>), [context-window](<https://devfeed.tech/tags/context-window.md>), [errors](<https://devfeed.tech/tags/errors.md>), [eval](<https://devfeed.tech/tags/eval.md>), [java](<https://devfeed.tech/tags/java.md>), [javascript](<https://devfeed.tech/tags/javascript.md>), [latency](<https://devfeed.tech/tags/latency.md>), [models](<https://devfeed.tech/tags/models.md>), [open](<https://devfeed.tech/tags/open.md>), [python](<https://devfeed.tech/tags/python.md>)

### AI overview

Mistral introduces Codestral, a 22B open-weight generative AI model designed for code generation. It supports code completion, fill-in-the-middle generation, function writing, and test writing across more than 80 programming languages. The article describes its 32k context window, benchmark performance, latency characteristics, and research/testing license.

### Source excerpt

The most powerful AI platform for enterprises. Customize, fine-tune, and deploy AI assistants, autonomous agents, and multimodal AI with open models.

## CodeGemma - an official Google release for code LLMs

DevFeed: [CodeGemma - an official Google release for code LLMs](<https://devfeed.tech/articles/codegemma-an-official-google-release-for-code-llms-7145.md>)

Original publisher: [Read original article](<https://huggingface.co/blog/codegemma>)

Author: Pedro Cuenca; Omar Sanseviero; Vaibhav Srivastav; Philipp Schmid; Mishig ᠮᠢᠰᠾᠢᠭ; Loubna Ben Allal

Published: 2024-04-09T00:00:00Z

Content type: release

Language: en

Sources: [Hugging Face - Blog](<https://devfeed.tech/sources/hugging-face-blog.md>)

Topics: [gemma](<https://devfeed.tech/topics/gemma.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>), [Google](<https://devfeed.tech/topics/google.md>), [hugging face](<https://devfeed.tech/topics/hugging-face.md>), [AI Development](<https://devfeed.tech/topics/ai-development.md>), [Transformers](<https://devfeed.tech/topics/transformers.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [Benchmark](<https://devfeed.tech/topics/benchmark.md>), [inference-endpoints](<https://devfeed.tech/topics/inference-endpoints.md>), [Cloud](<https://devfeed.tech/topics/cloud.md>), [Java](<https://devfeed.tech/topics/java.md>)

Tags: [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [c-plus-plus](<https://devfeed.tech/tags/c-plus-plus.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [code-completion](<https://devfeed.tech/tags/code-completion.md>), [community](<https://devfeed.tech/tags/community.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [gcp](<https://devfeed.tech/tags/gcp.md>), [gemma](<https://devfeed.tech/tags/gemma.md>), [google](<https://devfeed.tech/tags/google.md>), [hugging-face](<https://devfeed.tech/tags/hugging-face.md>), [inference-endpoints](<https://devfeed.tech/tags/inference-endpoints.md>), [java](<https://devfeed.tech/tags/java.md>), [javascript](<https://devfeed.tech/tags/javascript.md>), [llm](<https://devfeed.tech/tags/llm.md>), [llms](<https://devfeed.tech/tags/llms.md>), [nlp](<https://devfeed.tech/tags/nlp.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [research](<https://devfeed.tech/tags/research.md>), [transformers](<https://devfeed.tech/tags/transformers.md>)

### AI overview

Google's CodeGemma release introduces open-access 2B and 7B code-specialist LLMs, including base and instruction-tuned variants. The models support code infilling, completion, generation, code understanding, conversational use, and mathematical reasoning, with integrations across the Hugging Face ecosystem and Google Cloud.

### Source excerpt

We're on a journey to advance and democratize artificial intelligence through open source and open science.

## Kotlin 1.9.20: Streamlining Source Sets in Multiplatform Project - Jigar Brahmbhatt

DevFeed: [Kotlin 1.9.20: Streamlining Source Sets in Multiplatform Project - Jigar Brahmbhatt](<https://devfeed.tech/articles/kotlin-1-9-20-streamlining-source-sets-in-multiplatform-project-jigar-brahmbhatt-38259.md>)

Original publisher: [Read original article](<https://touchlab.co/kotlin-1-9-20-source-set-enhancements>)

Published: 2024-02-01T00:00:00Z

Content type: article

Language: en

Sources: [Touchlab | Enterprise Mobile Innovation & Development](<https://devfeed.tech/sources/touchlab-enterprise-mobile-innovation-development.md>)

Topics: [Kotlin](<https://devfeed.tech/topics/kotlin.md>), [Kotlin Multiplatform](<https://devfeed.tech/topics/kotlin-multiplatform.md>), [multiplatform](<https://devfeed.tech/topics/multiplatform.md>), [gradle-plugin](<https://devfeed.tech/topics/gradle-plugin.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>), [Code](<https://devfeed.tech/topics/code.md>), [Development](<https://devfeed.tech/topics/development.md>)

Tags: [boilerplate](<https://devfeed.tech/tags/boilerplate.md>), [code-completion](<https://devfeed.tech/tags/code-completion.md>), [development](<https://devfeed.tech/tags/development.md>), [gradle](<https://devfeed.tech/tags/gradle.md>), [gradle-plugin](<https://devfeed.tech/tags/gradle-plugin.md>), [kotlin](<https://devfeed.tech/tags/kotlin.md>), [kotlin-multiplatform](<https://devfeed.tech/tags/kotlin-multiplatform.md>), [multiplatform](<https://devfeed.tech/tags/multiplatform.md>), [new-features](<https://devfeed.tech/tags/new-features.md>)

### AI overview

This article explains how Kotlin 1.9.20 improves Kotlin Multiplatform source set configuration. It covers the default source set hierarchy template, simplified iOS target setup, custom source sets, manual template application, and enhanced IDE code completion.

### Source excerpt

Explore how Kotlin 1.9.20's new features significantly enhance the experience of setting up target source sets

## Autocomplete for Behaviors in PhpStorm with Mixins & Veneers

DevFeed: [Autocomplete for Behaviors in PhpStorm with Mixins & Veneers](<https://devfeed.tech/articles/autocomplete-for-behaviors-in-phpstorm-with-mixins-veneers-31244.md>)

Original publisher: [Read original article](<https://nystudio107.com/blog/autocomplete-for-behaviors-in-phpstorm-with-mixins-veneers>)

Author: andrew@nystudio107.com (Andrew Welch)

Published: 2022-12-20T15:42:00Z

Content type: tutorial

Language: en

Sources: [nystudio107 | Articles on modern web development.](<https://devfeed.tech/sources/nystudio107-articles-on-modern-web-development.md>)

Topics: [phpstorm](<https://devfeed.tech/topics/phpstorm.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>), [Content Management System](<https://devfeed.tech/topics/cms.md>), [Development](<https://devfeed.tech/topics/development.md>)

Tags: [added](<https://devfeed.tech/tags/added.md>), [autocomplete](<https://devfeed.tech/tags/autocomplete.md>), [behaviors](<https://devfeed.tech/tags/behaviors.md>), [cms](<https://devfeed.tech/tags/cms.md>), [code](<https://devfeed.tech/tags/code.md>), [code-analysis](<https://devfeed.tech/tags/code-analysis.md>), [code-completion](<https://devfeed.tech/tags/code-completion.md>), [completion](<https://devfeed.tech/tags/completion.md>), [craft](<https://devfeed.tech/tags/craft.md>), [insights](<https://devfeed.tech/tags/insights.md>), [intellisense](<https://devfeed.tech/tags/intellisense.md>), [learn](<https://devfeed.tech/tags/learn.md>), [mixins](<https://devfeed.tech/tags/mixins.md>), [phpstorm](<https://devfeed.tech/tags/phpstorm.md>), [project](<https://devfeed.tech/tags/project.md>), [using](<https://devfeed.tech/tags/using.md>), [veneers](<https://devfeed.tech/tags/veneers.md>)

### AI overview

A tutorial on restoring PhpStorm IntelliSense code completion for behaviors added to Craft CMS projects. It explains how to use mixins and veneer classes so PhpStorm can recognize dynamically added properties and methods through static code analysis.

### Source excerpt

Learn how to get PhpStorm IntelliSense code completion for Behaviors added to your Craft CMS project using Mixins & Veneers

## Maestro: The Orchestration Language Powering Shopify Flow

DevFeed: [Maestro: The Orchestration Language Powering Shopify Flow](<https://devfeed.tech/articles/maestro-the-orchestration-language-powering-shopify-flow-1474.md>)

Original publisher: [Read original article](<https://shopify.engineering/maestro-the-orchestration-language-powering-shopify-flow>)

Author: Thiago Tonelli

Published: 2022-05-13T18:45:00Z

Content type: article

Language: en

Sources: [Shopify Engineering](<https://devfeed.tech/sources/shopify-engineering.md>), [Shopify Engineering - Shopify Engineering](<https://devfeed.tech/sources/shopify-engineering-shopify-engineering.md>)

Topics: [Orchestration](<https://devfeed.tech/topics/orchestration.md>), [Shopify](<https://devfeed.tech/topics/shopify.md>), [Automation](<https://devfeed.tech/topics/automation.md>), [Data structures](<https://devfeed.tech/topics/data-structures.md>), [Usability](<https://devfeed.tech/topics/usability.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>)

Tags: [automation](<https://devfeed.tech/tags/automation.md>), [code-completion](<https://devfeed.tech/tags/code-completion.md>), [data](<https://devfeed.tech/tags/data.md>), [orchestration](<https://devfeed.tech/tags/orchestration.md>), [scale](<https://devfeed.tech/tags/scale.md>), [shopify](<https://devfeed.tech/tags/shopify.md>), [usability](<https://devfeed.tech/tags/usability.md>), [use-cases](<https://devfeed.tech/tags/use-cases.md>)

### AI overview

Shopify Engineering describes Maestro, a domain-specific orchestration language created to support the Flow engine's scalability, fault tolerance, and execution requirements. The article explains how Shopify decoupled the visual Flow editor from its execution engine, introduced a horizontally scalable Maestro engine, and migrated existing workflows through a translation layer.

### Source excerpt

An overview of Shopify Flow's new orchestration language Maestro, highlighting its design and implementation as well as how neatly it integrates with and addresses the requirements.

## Code Completion. Часть 4: обучение модели

DevFeed: [Code Completion. Часть 4: обучение модели](<https://devfeed.tech/articles/code-completion-4-23939.md>)

Original publisher: [Read original article](<https://habr.com/ru/companies/JetBrains/articles/582388/>)

Author: p0b0rchy (JetBrains)

Published: 2021-10-11T09:02:36Z

Content type: article

Language: ru

Sources: [JetBrains RU](<https://devfeed.tech/sources/jetbrains-ru.md>)

Topics: [code-completion](<https://devfeed.tech/topics/code-completion.md>), [машинное обучение](<https://devfeed.tech/topics/tag-055aee430837.md>), [ide](<https://devfeed.tech/topics/ide.md>)

Tags: [catboost](<https://devfeed.tech/tags/catboost.md>), [code-completion](<https://devfeed.tech/tags/code-completion.md>), [ide](<https://devfeed.tech/tags/ide.md>), [metric-learning](<https://devfeed.tech/tags/metric-learning.md>), [ml](<https://devfeed.tech/tags/ml.md>), [tag-055aee430837](<https://devfeed.tech/tags/tag-055aee430837.md>), [tag-baed6d151ab4](<https://devfeed.tech/tags/tag-baed6d151ab4.md>)

### AI overview

This fourth article in a code-completion series describes problems encountered while training a machine-learning model. It explains anonymizing collected IDE data, investigating anomalous autocomplete usage through logs, and excluding users with alphabetical suggestion sorting from the training dataset and experiments.

### Source excerpt

Хотим поделиться специфическими проблемами, возникшими у нас при обучении модели, и решениями, которые мы для этих проблем придумали. В прошлый раз мы остановились на том, что из-за необходимости защищать код пользователей даже от себя самих нам необходимо анонимизировать все собранные данные. Вызову окна автодополнения сопоставляется набор параметров, каждый из которых -- либо число, либо перечислимый тип. А вот сам код, на котором пользователь вызвал окно, мы не сохраняем. Если при сборе данных мы допустим ошибку, нельзя будет просто заглянуть в код и все понять. Нам придется устраивать целое расследование без всякой гарантии успеха. Например

## Code Completion. Часть 3: где взять данные для обучения?

DevFeed: [Code Completion. Часть 3: где взять данные для обучения?](<https://devfeed.tech/articles/code-completion-3-23935.md>)

Original publisher: [Read original article](<https://habr.com/ru/companies/JetBrains/articles/578650/>)

Author: p0b0rchy (JetBrains)

Published: 2021-09-20T10:27:02Z

Content type: article

Language: ru

Sources: [JetBrains RU](<https://devfeed.tech/sources/jetbrains-ru.md>)

Topics: [Code](<https://devfeed.tech/topics/code.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>), [машинное обучение](<https://devfeed.tech/topics/tag-055aee430837.md>), [jetbrains](<https://devfeed.tech/topics/jetbrains.md>)

Tags: [code](<https://devfeed.tech/tags/code.md>), [code-completion](<https://devfeed.tech/tags/code-completion.md>), [data-protection](<https://devfeed.tech/tags/data-protection.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [ide](<https://devfeed.tech/tags/ide.md>), [jetbrains](<https://devfeed.tech/tags/jetbrains.md>), [tag-055aee430837](<https://devfeed.tech/tags/tag-055aee430837.md>), [tag-9bf5e01ce62e](<https://devfeed.tech/tags/tag-9bf5e01ce62e.md>), [tag-baed6d151ab4](<https://devfeed.tech/tags/tag-baed6d151ab4.md>)

### AI overview

This article explains how JetBrains approaches building training data for machine-learning-based code completion. It describes why user code cannot be collected and why datasets derived only from completed open-source repositories do not adequately represent real development scenarios.

### Source excerpt

В предыдущей статье цикла мы установили, что современной системе автодополнения нужно машинное обучение -- чтобы ранжировать варианты подсказок. Машине, как и человеку, для обучения нужны данные. Мы используем подход "обучение с учителем". Он предполагает, что мы показываем алгоритму примеры задач, для которых нам заранее известно правильное решение. Алгоритм извлекает из этих примеров закономерности и учится принимать правильные решения даже в тех ситуациях, которых он раньше не видел. Результат обучения критическим образом зависит от качества и объема этих примеров. И именно здесь у нас возникают трудности. С одной стороны, собирать базу на основе кода наших пользователей нельзя: результат вашего труда остается только у вас, на серверы JetBrains он не попадает. С другой стороны, из доступных репозиториев с открытым кодом полноценный датасет собрать не получается. Делая так, мы обучали бы алгоритм работать только с готовым (рабочим) кодом. Сегодня обсудим, как мы выкручиваемся из этой ситуации. Для начала давайте более подробно рассмотрим, почему не получается работать с готовыми репозиториями. А потому что они "готовые"

## Code Completion. Часть 2: зачем тут машинное обучение?

DevFeed: [Code Completion. Часть 2: зачем тут машинное обучение?](<https://devfeed.tech/articles/code-completion-2-23929.md>)

Original publisher: [Read original article](<https://habr.com/ru/companies/JetBrains/articles/572864/>)

Author: p0b0rchy (JetBrains)

Published: 2021-08-17T09:45:16Z

Content type: tutorial

Language: ru

Sources: [JetBrains RU](<https://devfeed.tech/sources/jetbrains-ru.md>)

Topics: [code-completion](<https://devfeed.tech/topics/code-completion.md>), [машинное обучение](<https://devfeed.tech/topics/tag-055aee430837.md>), [ide](<https://devfeed.tech/topics/ide.md>), [Code](<https://devfeed.tech/topics/code.md>)

Tags: [code](<https://devfeed.tech/tags/code.md>), [code-completion](<https://devfeed.tech/tags/code-completion.md>), [ide](<https://devfeed.tech/tags/ide.md>), [tag-055aee430837](<https://devfeed.tech/tags/tag-055aee430837.md>), [tag-baed6d151ab4](<https://devfeed.tech/tags/tag-baed6d151ab4.md>)

### AI overview

The second part of a series explains why machine learning is used in code completion. It examines conflicts among heuristic rules, including prefix and CamelCase matching, and discusses ranking suggestions by correctness and proximity to the edited code.

### Source excerpt

В первой части мы сформулировали, из каких компонентов состоит система автодополнения, обсудили способы ее использования и требования к качеству. Теперь давайте разберемся, зачем там нужно машинное обучение. Казалось бы, страшно выбрасывать работающий код и заменять его машиннообученным бинарником, который жрет память, может замедлить работу IDE, да еще не вдруг и отладишь его, если что-то пошло не так. В нашем случае "работающий код" -- это эвристики, жесткие правила. Они отлично работают, пока их не очень много и они не конфликтуют между собой. Давайте рассмотрим такие правила подробнее и разберемся, откуда берутся конфликты. Как на самом деле работают подсказки

## Code Completion. Часть 1: сценарии и требования

DevFeed: [Code Completion. Часть 1: сценарии и требования](<https://devfeed.tech/articles/code-completion-1-23920.md>)

Original publisher: [Read original article](<https://habr.com/ru/companies/JetBrains/articles/568884/>)

Author: p0b0rchy (JetBrains)

Published: 2021-07-21T11:18:54Z

Content type: article

Language: ru

Sources: [JetBrains RU](<https://devfeed.tech/sources/jetbrains-ru.md>)

Topics: [code-completion](<https://devfeed.tech/topics/code-completion.md>), [Code](<https://devfeed.tech/topics/code.md>), [ide](<https://devfeed.tech/topics/ide.md>), [машинное обучение](<https://devfeed.tech/topics/tag-055aee430837.md>)

Tags: [code](<https://devfeed.tech/tags/code.md>), [code-completion](<https://devfeed.tech/tags/code-completion.md>), [ide](<https://devfeed.tech/tags/ide.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [tag-055aee430837](<https://devfeed.tech/tags/tag-055aee430837.md>)

### AI overview

This first article in a series examines why machine learning is useful for code completion in JetBrains IDEs. It introduces completion-use scenarios and requirements, including dataset collection, data protection, model training, and quality evaluation with desktop-application A/B tests.

### Source excerpt

Знаете, сколько всего успевает сделать IDE, чтобы показать окно сode сompletion, когда вы начинаете набирать новое слово? Под капотом формулы, полученные с помощью машинного обучения. Сортировка подсказок -- одно из первых применений машинного обучения в IDE JetBrains. Думаем, вам будет интересно узнать, через какие тернии нам пришлось пройти, чтобы эту сортировку реализовать. Данная статья -- первая из цикла, в рамках которого мы расскажем, зачем окну с подсказками вообще нужно машинное обучение и как мы решали некоторые связанные с этим задачи. А задач тех немало

## Introducing Tailwind Play

DevFeed: [Introducing Tailwind Play](<https://devfeed.tech/articles/introducing-tailwind-play-3474.md>)

Original publisher: [Read original article](<https://tailwindcss.com/blog/introducing-tailwind-play>)

Published: 2020-10-07T13:00:00Z

Content type: article

Language: en

Sources: [Tailwind CSS Blog](<https://devfeed.tech/sources/tailwind-css-blog.md>)

Topics: [Tailwind CSS](<https://devfeed.tech/topics/tailwind.md>), [browser](<https://devfeed.tech/topics/browser.md>), [build performance](<https://devfeed.tech/topics/build-performance.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>), [PostCSS](<https://devfeed.tech/topics/postcss.md>), [Responsive Design](<https://devfeed.tech/topics/responsive-design.md>), [real-time](<https://devfeed.tech/topics/real-time.md>)

Tags: [browser](<https://devfeed.tech/tags/browser.md>), [build](<https://devfeed.tech/tags/build.md>), [chrome](<https://devfeed.tech/tags/chrome.md>), [code-completion](<https://devfeed.tech/tags/code-completion.md>), [playground](<https://devfeed.tech/tags/playground.md>), [real-time](<https://devfeed.tech/tags/real-time.md>), [responsive-design](<https://devfeed.tech/tags/responsive-design.md>), [tailwind](<https://devfeed.tech/tags/tailwind.md>)

### AI overview

Tailwind Play is an online playground that brings Tailwind CSS build-time features into the browser. It supports theme customization, variants, custom directives, plugins, code completion, linting, responsive design mode, real-time previews, and one-click sharing.

### Source excerpt

Tailwind Play is an advanced online playground for Tailwind CSS that lets you use all of Tailwind's build-time features directly in the browser.

## Temporal Transparency Update #5

DevFeed: [Temporal Transparency Update #5](<https://devfeed.tech/articles/temporal-transparency-update-5-36038.md>)

Original publisher: [Read original article](<https://temporal.io/blog/temporal-transparency-5>)

Author: Ryland Goldstein

Published: 2020-08-07T04:00:00Z

Content type: release

Language: en

Sources: [Temporal Blog](<https://devfeed.tech/sources/temporal-blog.md>)

Topics: [SDKs](<https://devfeed.tech/topics/sdks.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>), [Refactoring](<https://devfeed.tech/topics/refactoring.md>), [releases](<https://devfeed.tech/topics/releases.md>), [schema-evolution](<https://devfeed.tech/topics/schema-evolution.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [Apache Cassandra](<https://devfeed.tech/topics/cassandra.md>), [MySQL](<https://devfeed.tech/topics/mysql.md>)

Tags: [cassandra](<https://devfeed.tech/tags/cassandra.md>), [ci-cd](<https://devfeed.tech/tags/ci-cd.md>), [code-completion](<https://devfeed.tech/tags/code-completion.md>), [compatibility](<https://devfeed.tech/tags/compatibility.md>), [mysql](<https://devfeed.tech/tags/mysql.md>), [product-news](<https://devfeed.tech/tags/product-news.md>), [refactoring](<https://devfeed.tech/tags/refactoring.md>), [releases](<https://devfeed.tech/tags/releases.md>), [schema](<https://devfeed.tech/tags/schema.md>), [update](<https://devfeed.tech/tags/update.md>)

### AI overview

Temporal's fifth transparency update reports that JavaSDK bugs were addressed through a major refactor now under review. The update says the JavaSDK will become code complete after review, while the team focuses on stabilization, compatibility, and automated CI/CD pipelines ahead of the production V1 release.

### Source excerpt

This was a crazy week for us in a good way. We harnessed the momentum of our code completion efforts to stay extra focused on our next steps. In this update we're going to talk about the JavaSDK, stabilization, and more.

## Temporal Transparency Update #3

DevFeed: [Temporal Transparency Update #3](<https://devfeed.tech/articles/temporal-transparency-update-3-36080.md>)

Original publisher: [Read original article](<https://temporal.io/blog/transparency-update-3>)

Author: Ryland Goldstein

Published: 2020-07-25T04:00:00Z

Content type: news

Language: en

Sources: [Temporal Blog](<https://devfeed.tech/sources/temporal-blog.md>)

Topics: [code-completion](<https://devfeed.tech/topics/code-completion.md>), [Code](<https://devfeed.tech/topics/code.md>)

Tags: [code](<https://devfeed.tech/tags/code.md>), [code-completion](<https://devfeed.tech/tags/code-completion.md>), [completion](<https://devfeed.tech/tags/completion.md>), [product-news](<https://devfeed.tech/tags/product-news.md>), [update](<https://devfeed.tech/tags/update.md>)

### AI overview

Temporal's third transparency update for V1 code completion reports that the July 31, 2020 completion date remained unchanged, about half of the remaining tasks were completed, and the team expected to move toward stabilization as work finished.

### Source excerpt

Hope everyone had a productive and enjoyable week! I'm here to share our third update for V1 code completion.

## Fixing autocompletion on mixed Objective-C and Swift projects

DevFeed: [Fixing autocompletion on mixed Objective-C and Swift projects](<https://devfeed.tech/articles/fixing-autocompletion-on-mixed-objective-c-and-swift-projects-21621.md>)

Original publisher: [Read original article](<http://miqu.me/blog/2017/06/16/fixing-autocompletion-on-mixed-objective-c-and-swift-projects/>)

Author: Miguel Angel Quiñones

Published: 2017-06-16T22:55:27Z

Content type: tutorial

Language: en

Sources: [Miguel Quinones](<https://devfeed.tech/sources/miguel-quinones.md>)

Topics: [Swift](<https://devfeed.tech/topics/swift.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>), [Xcode](<https://devfeed.tech/topics/xcode.md>), [Objective-C](<https://devfeed.tech/topics/objective-c.md>), [debugging](<https://devfeed.tech/topics/debugging.md>), [Library](<https://devfeed.tech/topics/library.md>)

Tags: [code-completion](<https://devfeed.tech/tags/code-completion.md>), [debug](<https://devfeed.tech/tags/debug.md>), [debugging](<https://devfeed.tech/tags/debugging.md>), [latest-release](<https://devfeed.tech/tags/latest-release.md>), [libraries](<https://devfeed.tech/tags/libraries.md>), [objective-c](<https://devfeed.tech/tags/objective-c.md>), [swift](<https://devfeed.tech/tags/swift.md>), [xcode](<https://devfeed.tech/tags/xcode.md>)

### AI overview

This article explains intermittent Swift code-completion failures in mixed Objective-C and Swift projects using Xcode 8 or Xcode 9 beta. It attributes the problem to library classes imported without their framework or library path, which can prevent SourceKit from indexing correctly. The article recommends including the path in bridging-header imports and describes using Xcode's debug mode and SourceKit service logs to investigate the issue. An update says the problem appeared to be fixed in Xcode 9.0.1.

### Source excerpt

This year I had the privilege of attending WWDC for the first time. I knew the labs were very important, and I want to share a resolution to fix a problem for our big project at Peak: Autocompletion on Swift code was not working most of the time in the IDE, and with the help of an Apple engineer in the labs we got it sorted. Update on 30/10/2017: The issues described here seem to be fixed with latest release version of Xcode 9.0.1 (9A1004). Also the 'internal' menu doesn't seem to be accessible anymore. If you know how to access it I'd love to hear from you! The issue If you are using Xcode 8 or Xcode 9 (beta 1 as of this writing), when you have a project with mixed Objective-C and Swift, you might encounters problems with autocompletion inside the Swift code. We've had this intermitently and most recent IDE versions made work very difficult as we would not have any autocompletion for our Swift code, or it would work intermitently. Our gut feeling was that the bridging header had some classes that were causing SourceKit to fail, but we didn't know how to debug the issue nor provide good feedback for Apple to fix it. Our chance came while visiting the labs at the conference, and a very helpful and patient Apple engineer worked with our project to understand what was wrong. Bridging header imports It turns out that if you import classes from libraries (in our case using CocoaPods) omitting the path may result in Sourcekit unable to index properly. For example, a class in some library: #import "MyClass.h" Code will compile fine but your code completion in Swift side might break because the indexer fails when encountering this file. I'm still unsure if it's because of importing the file somewhere else differently or if it's just a bug in the indexer. Anyway, the way to work around it (Apple is supposedly aware of this) is to import with the framework or library style, including the path: #import "path-to-library/MyClass.h" How to debug The engineer kindly made a wrote u

## IDEs and Productivity

DevFeed: [IDEs and Productivity](<https://devfeed.tech/articles/ides-and-productivity-24947.md>)

Original publisher: [Read original article](<https://codeahoy.com/2016/06/10/ides-and-developer-productivity/>)

Author: umer

Published: 2016-06-10T00:00:00Z

Content type: opinion

Language: en

Sources: [Code Ahoy - Articles](<https://devfeed.tech/sources/code-ahoy-articles.md>)

Topics: [ide](<https://devfeed.tech/topics/ide.md>), [code productivity](<https://devfeed.tech/topics/code-productivity.md>), [Vim](<https://devfeed.tech/topics/vim.md>), [Emacs](<https://devfeed.tech/topics/emacs.md>), [Refactoring](<https://devfeed.tech/topics/refactoring.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>), [Maven](<https://devfeed.tech/topics/maven.md>)

Tags: [code-completion](<https://devfeed.tech/tags/code-completion.md>), [developer-productivity](<https://devfeed.tech/tags/developer-productivity.md>), [emacs](<https://devfeed.tech/tags/emacs.md>), [ide](<https://devfeed.tech/tags/ide.md>), [productivity](<https://devfeed.tech/tags/productivity.md>), [refactoring](<https://devfeed.tech/tags/refactoring.md>), [vim](<https://devfeed.tech/tags/vim.md>)

### AI overview

This opinion article argues that learning an IDE's features can improve developer productivity, particularly in complex enterprise applications. It describes benefits including debugging, dependency-graph analysis, refactoring, code completion, and boilerplate generation, while acknowledging that experienced Vim or Emacs users may already be productive with their chosen editor.

### Source excerpt

I used to be neutral on the choice and even the use of an IDE for writing code. In university, I learned and used Vim for assignments. When I started my first job, I switched to a popular Java IDE because everyone at work was using it and it featured a nice debugger. Other than debugging and basic auto-completion, I didn't learn the IDE much in terms of its features. But that changed one day when I ran into an issue compiling a Java project. I went over to the developer's desk who had made the most recent commit. He opened up the project in his IDE and in just a few keystrokes, he had the entire maven dependency graph displayed on his screen. Just like magic. Few more keystrokes and he was able to locate the root cause. I was surprised because after months of using the IDE, I had no idea it had that feature. Two things happened: I realized that up until that point, I was using the IDE like a text editor and missing out on many of its powerful features. I became convinced that knowing how to effectively use an IDE has a profound impact on developer productivity. This is especially true for complex enterprise applications, where superior knowledge of IDE can provide a significant boost in productivity. I started out with Vim and loved it. When I first tried Eclipse/NetBeans for Java development, I was overwhelmed by the number things that were cluttered on the screen. The second thing that put me off was that I had to use the mouse a lot for navigation. But I saw the value in refactoring, code-completion and generating boiler-plate code so I decided to bite the bullet and switched to it from Vim. In retrospect, it was a good decision. Many developers who become proficient in text editors like Vim or Emacs don't switch to an IDE. I worked with a developer who wrote code exclusively in Emacs. When I hired him, he told that he'll use nothing but Emacs. I was sure that in time I'll be able to 'show him the light' and that he'll see the value in using an IDE. But after wat

## Android Studio Tips and Tricks

DevFeed: [Android Studio Tips and Tricks](<https://devfeed.tech/articles/android-studio-tips-and-tricks-25102.md>)

Original publisher: [Read original article](<http://michaelevans.org/blog/2016/01/06/android-studio-tips-and-tricks/>)

Author: Michael Evans

Published: 2016-01-07T01:25:53Z

Content type: tutorial

Language: en

Sources: [Gadget Habit](<https://devfeed.tech/sources/gadget-habit.md>)

Topics: [Android Studio](<https://devfeed.tech/topics/android-studio.md>), [IntelliJ IDEA](<https://devfeed.tech/topics/intellij-idea.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>), [JSON](<https://devfeed.tech/topics/json.md>), [Regular expression](<https://devfeed.tech/topics/regular-expression.md>), [API](<https://devfeed.tech/topics/api.md>)

Tags: [android](<https://devfeed.tech/tags/android.md>), [android-studio](<https://devfeed.tech/tags/android-studio.md>), [code-completion](<https://devfeed.tech/tags/code-completion.md>), [development](<https://devfeed.tech/tags/development.md>), [json](<https://devfeed.tech/tags/json.md>), [productivity](<https://devfeed.tech/tags/productivity.md>), [shortcuts](<https://devfeed.tech/tags/shortcuts.md>), [tips-and-tricks](<https://devfeed.tech/tags/tips-and-tricks.md>)

### AI overview

A practical guide to Android Studio and IntelliJ features that improve coding productivity, including language injection for JSON and regular expressions, context-aware type completion, the Productivity Guide, and IntelliJ's REST client.

### Source excerpt

I recently attended Google's Android Dev Summit where the Tools team presented a talk entitled Android Studio For Experts. The room was packed for the 90 minute session, where a lot of great Android Studio tips were shared. This gave me the idea of showing off some of my favorite Android Studio tips! Language Injection Ever needed to type a JSON String? Perhaps you've used one as a text fixture for one of your GSON deserializers and know that it's a huge pain to manage all those backslashes. Fortunately, IntelliJ has a feature called Language Injection, which allows you to edit the JSON fragment in its own editor, and then IntelliJ will properly inject that fragment into your code as an escaped String. Inject Language/Reference is an intention action1, so you can start it by using ⌥+Return, or ⌘+⇧+A and searching for it. Check RegExp This is pretty similar to the last tip, but if you select the language of the fragment as "RegExp", you'll get a handy regular expression tester! Smart(er) Completion Now I'm pretty sure most of you have used IntelliJ's code completion features. Press ⌥+Space, and IntelliJ/Android Studio lists options to complete the names of classes, methods, fields, and keywords within the visibility scope. But have you ever noticed that the suggestions seem to be based off the characters you've typed, rather than the actual types that are expected in the scope of the caret? Something like this: Well if you use Type Completion (by pressing ⌥+⇧+Space), you will see a list of suggestions containing only those types that are applicable to the current context. In the example below, you'll only get types that return a Reader, which is the type that the BufferedReader's constructor expects: What's even cooler is that you can press it an additional time, and IntelliJ will do a deeper scan (looking at static method calls, chained expressions, etc.) to find more options for you: Discovering Your Own Tips and Tricks Another really cool feature is the Productivi

## Choosing Atom as a Code Editor for Python, Perl, and Ruby

DevFeed: [Choosing Atom as a Code Editor for Python, Perl, and Ruby](<https://devfeed.tech/articles/searching-for-better-code-editor-27608.md>)

Original publisher: [Read original article](<https://gagor.pro/2014/01/searching-for-better-code-editor/>)

Author: Tom

Published: 2014-01-24T00:00:00Z

Content type: opinion

Language: en

Sources: [Tomasz Gągor](<https://devfeed.tech/sources/tomasz-gagor.md>)

Topics: [Atom](<https://devfeed.tech/topics/atom.md>), [Sublime Text](<https://devfeed.tech/topics/sublime-text.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>), [Syntax Highlighting](<https://devfeed.tech/topics/syntax-highlighting.md>)

Tags: [autocomplete](<https://devfeed.tech/tags/autocomplete.md>), [syntax-highlighting](<https://devfeed.tech/tags/syntax-highlighting.md>)

### AI overview

The author compares several code editors for work involving Python, Perl, Ruby, and related frameworks. After trying Brackets, they switch to Atom because it better fits their needs and list plugins and themes they use.

### Source excerpt

I've been using different code editors for different purposes. Gedit was fine for small scripts but not for bigger projects. It lacks intelligent code completion (function/class names, etc.). I was searching for convenient editor for Python, Perl, Ruby with support for frameworks like Django, Rails, etc. I know Sublime Text - but it's paid external link . There is LimeText external link - open source clone, but it's not ready to be used on daily basics.

## Scala Unicode Arrows in IntelliJ IDEA

DevFeed: [Scala Unicode Arrows in IntelliJ IDEA](<https://devfeed.tech/articles/scala-unicode-arrows-in-intellij-idea-20848.md>)

Original publisher: [Read original article](<http://themodernlife.net/scala/intellij/2014/01/08/scala-unicode-arrows-in-intellij-idea/>)

Published: 2014-01-08T11:50:13Z

Content type: tutorial

Language: en

Sources: [Ian Hummel](<https://devfeed.tech/sources/ian-hummel.md>)

Topics: [IntelliJ IDEA](<https://devfeed.tech/topics/intellij-idea.md>), [Scala](<https://devfeed.tech/topics/scala.md>), [code-completion](<https://devfeed.tech/topics/code-completion.md>), [Sublime Text](<https://devfeed.tech/topics/sublime-text.md>)

Tags: [code-completion](<https://devfeed.tech/tags/code-completion.md>), [coding](<https://devfeed.tech/tags/coding.md>), [github](<https://devfeed.tech/tags/github.md>), [idea](<https://devfeed.tech/tags/idea.md>), [intellij](<https://devfeed.tech/tags/intellij.md>), [intellij-idea](<https://devfeed.tech/tags/intellij-idea.md>), [scala](<https://devfeed.tech/tags/scala.md>)

### AI overview

A tutorial explains how to use IntelliJ IDEA Live Templates to convert typed Scala operators such as =>, ->, and <- into Unicode arrows. It also contrasts this workflow with a Sublime Text project that provides Scala tab completion.

### Source excerpt

Several of my colleagues love IntelliJ for coding in Scala. I was pretty happy with Sublime Text 2 (and still use it for Ruby/Python/Shell/whatever) but the lack of code completion was really starting to affect my productivity. I spent way too much time looping through the edit/compile/fix typo cycle. Before I could switch though, I really wanted my fancy arrows in Scala! I have a GitHub project which adds tab completion in Sublime Text for Scala to turn "=>" into "⇒", "->" into "->" and "<-" into "<-". Turns out this is really easy to accomplish in IntelliJ as well! You just need to create a few Live Templates The Abbreviation field should be set to "=>" and the Template text field should be "⇒ " (I left a trailing space so my cursor position gets updated). Repeat for any other operators. Also ensure that you change the applicable context to Scala. Now when you type "=>" and hit you should get a unicode arrow, "⇒"!