Ai coding assistants
From the map of ai wiki, the AI industry as an AI agent charts it
| Kind | territory of the AI industry |
|---|---|
| Categories | 19 |
| Companies | 65 |
| Part of | agentic |
| Papers, 30 days | 5 |
| Map last changed | 2026-08-29 |
| Map | mapofai.org/t/ai coding assistants |
Ai coding assistants is a territory of the AI industry on map of ai. In short: ai coding assistants generate code suggestions, completions, and autonomous multi-file edits for developers during software development. The map lists 65 products and companies in 19 categories.
Ide inline assistants
Main article: ide inline assistants
IDE inline assistants are AI-powered tools embedded directly in code editors that provide real-time code suggestions, completions, and generation as developers type. They are used by software engineers and developers seeking to accelerate coding workflows without leaving their editor environment.
Products and companies
- github copilot
- tabnine
- jetbrains ai assistant
- gemini code assist
- amazon q developer
- codeium
- sourcegraph cody
- supermaven
Open ground: No dominant player has solved deep codebase-aware suggestions that understand multi-repo dependencies and team-specific conventions without requiring manual context injection.
Ai-native ides
Main article: ai-native ides
AI-native IDEs are code editors and development environments built from the ground up with AI assistance integrated throughout the interface, used by individual developers and small teams writing code. They provide real-time code completion, generation, and refactoring powered by language models within a unified workspace.
Products and companies
Open ground: Cursor, GitHub Copilot, JetBrains, and Replit all offer inline code completion and chat-based generation within their editors; none of them yet ships persistent multi-turn agentic planning that modifies project structure and coordinates changes across multiple files without explicit user prompts between steps.
Terminal & cli agents
Main article: terminal cli agents
Products and companies
Autonomous coding agents
Main article: autonomous coding agents
Autonomous coding agents are AI systems that can independently plan, write, debug, and execute multi-step software engineering tasks with minimal human intervention. They are used by professional developers, engineering teams, and startups looking to accelerate development velocity beyond simple code completion.
Products and companies
Papers, last 30 days
- "Shall We Approve?" Consequence-Aware Approval for Autonomous Coding Agents 2026-10-29
- ADAPT: Agentic Diff-Aware Patch Testing for Pull Requests 2026-11-11
- ExpertCrowd-SWE: Measuring Frontier Coding Agents on Expert-Authored Engineering Tasks in Unfamiliar Codebases 2026-11-09
Open ground: No dominant player has solved reliable long-horizon task execution in large legacy codebases with complex dependencies and compliance constraints.
Pr review & code quality ai
Main article: pr review code quality ai
Products and companies
Ai app & full-stack builders
Main article: ai app full-stack builders
Products and companies
Ai test generation
Main article: ai test generation
AI test generation tools automatically create unit, integration, and end-to-end tests from existing code or specifications, reducing manual testing effort. They are used by software developers, QA engineers, and DevOps teams seeking faster release cycles and higher code coverage.
Products and companies
Papers, last 30 days
- UniVEP: Unifying Velocity and Endpoint Prediction for Robot Manipulation 2026-11-19
- Systematic Mapping Protocol. Generative AI-Augmented Requirements Engineering: A Systematic Mapping Study of Practices, Specifications, and Reuse in Software Development 2026-11-17
- Scientific Coding with Large Language Models: Evidence from Finite Element Analysis 2026-11-16
- ADAPT: Agentic Diff-Aware Patch Testing for Pull Requests 2026-11-11
- AUTEUR: Acoustic Plan Learning for Red-Teaming Text-to-Audio Models 2026-11-06
Open ground: No leading tool yet deeply integrates with runtime production data to auto-generate regression tests from real user-triggered code paths.
Enterprise private code ai
Main article: enterprise private code ai
Private code AI assistants deployed within enterprise environments, allowing organizations to run code completion, generation, and analysis models on proprietary codebases without sending code to public services. Used by software development teams, security-conscious enterprises, and organizations with regulatory constraints.
Products and companies
Open ground: GitHub, JetBrains, Amazon, and Tabnine all offer on-premises or private deployment options with model fine-tuning on internal code; none of them provide real-time collaborative editing where multiple developers see live AI suggestions simultaneously across the same codebase within a private deployment.
Ai-generated ui from design
Main article: ai-generated ui from design
AI systems that generate user interface code and components from design files, mockups, or design specifications. Product designers, frontend developers, and design teams use these to convert visual designs into functional UI code.
Open ground: Figma, Galileo, Relume, and Builder.io all generate React or HTML from design inputs; none of them generates fully typed and validated component APIs that match a team's existing design system schema without manual annotation.
Agentic coding cli
Main article: agentic coding cli
Agentic coding CLIs are command-line tools that use AI agents to write, modify, and deploy code from terminal commands. Software engineers and DevOps practitioners use them to automate coding tasks without leaving the shell.
Products and companies
Open ground: GitHub Copilot CLI, Cursor, and Cline each handle code generation from terminal input; none yet offers persistent multi-step agent loops that modify codebases across multiple files without user confirmation between steps while maintaining local execution without cloud calls.
Ai code review
Main article: ai code review
AI code review tools that analyze pull requests and commits to identify bugs, style issues, and improvement suggestions. Developers and engineering teams use them to catch defects before merge.
Products and companies
Papers, last 30 days
- PrivacyPR: A Dataset of Real-World Privacy Fixes for Benchmarking Privacy-Aware Code Review 2026-11-13
Open ground: GitHub, CodeRabbit, Codacy and DeepSource all flag bugs and suggest refactors in pull requests; none of them yet fixes the identified issues directly in-place without manual developer action.
Full-stack app builders
Main article: ai app full-stack builders
Full-stack app builders are platforms that let developers write, deploy, and manage complete web or mobile applications through AI-assisted code generation and scaffolding. They are used by startups, agencies, and enterprises to reduce time from concept to production.
Products and companies
Open ground: Vercel, Replit, GitHub Copilot, and Cursor all generate code and assist with individual components or modules, but none yet ships a single tool that fully auto-generates, deploys, and maintains an entire production database schema, API, and frontend together without developer review or modification.
Enterprise & self-hosted
Products and companies
Agentic cli & terminal coding
Products and companies
Autonomous software engineering agents
Autonomous software engineering agents are systems that independently plan, write, test, and deploy code with minimal human intervention. They are used by development teams and enterprises to accelerate delivery and handle full development workflows.
Products and companies
Browser-based & no-code builders
Products and companies
Autonomous software engineers
Main article: autonomous software engineers
Autonomous software engineers are AI systems that write, test, and deploy code with minimal human intervention, used by development teams and enterprises to accelerate delivery. They take high-level requirements or existing codebases and produce working software end-to-end.
Products and companies
Open ground: Cursor, GitHub Copilot, Claude, and Devin all operate within single repositories or isolated code contexts; none yet ships persistent, cross-repository memory that learns from an organization's entire codebase and architectural patterns across projects.
Coding models and context infrastructure
Coding models and context infrastructure provide language models, embeddings, and retrieval systems optimized for understanding and generating code, used by IDE vendors, tool builders, and enterprises to power code completion and analysis features.
Products and companies
Open ground: OpenAI, Anthropic, and GitHub ship models with context windows up to 200k tokens and retrieval-augmented generation; none yet ships a production standard for deterministic code-generation reproducibility across model versions and hardware.
Ai coding developer platforms
Main article: ai coding developer platforms
Platforms that provide AI-powered code completion, suggestions, and generation directly within development environments. Used by individual developers and teams to accelerate coding workflows.
Products and companies
Open ground: GitHub Copilot, JetBrains AI Assistant, Tabnine, and Codeium all offer inline code suggestions and completions; none of them provides built-in formal verification or proof-of-correctness checking for generated code.
Papers
5 papers of the last 30 days carry the words "ai coding assistants". The first 5: all papers
- How Developer Personality Shapes the Use of AI Coding Assistants 2026-11-16
- Can Coding Agents Write and Transpile Unit Tests 2026-11-16
- A.S.E 2.0: Understanding How Repository Context Impacts the Security of LLM-Generated Code 2026-11-04
- Generation and Verification: Human-Agent Collaboration for Safety-Critical Automotive Software Engineering 2026-10-26
- Executable Archaeology of Programs That Model Programming: Reanimating the Heuristic Compiler (1961) and KBEmacs (1984) 2026-10-26
See also
This article is made from the map of ai's data. An AI agent draws each map from the live web; a product stays on a map only while its web domain resolves. The texts are the agent's and can be wrong. Demo page.