AI Directory · Researched reviews · ASEAN lens
Code & Dev Tools
136 tools · Code & Dev Tools · Most popular
How to choose an AI coding tool
Benchmark scores are the least useful signal here. They measure self-contained puzzles; you are buying help with a codebase that has history, conventions and a test suite someone else wrote.
Split them by how much of your code they see. Autocomplete works from the open file and a little context around it: fast, cheap, and it never surprises you. An agent reads the repository, edits several files, and runs commands, which is far more useful and far more capable of being confidently wrong. Most teams end up wanting both, so the honest question is whether you will review what the agent writes.
Check where a tool runs before you check what it can do. A tool that requires uploading your repository is a different procurement conversation from one that runs locally or inside your own cloud. For regulated work, that difference decides it outright.
Then check the exit. Editor forks, IDE plugins, and CLI tools carry very different switching costs, and this category consolidates fast; several of the tools listed here have changed owners in the past year. Prefer tools whose output is ordinary code in an ordinary repo.
Also listed in Code & Dev Tools
47 more tools we track but have not written up in depth. Each entry links to the vendor's own site; we have not tested these tools.
-
Aider Open-source terminal pair programmer that commits through Git, now stalled
-
AskCodi AI coding cofounder and multi-model gateway — a free million tokens a month, Pro from $29
-
BAML A typed programming language for agent code, with LLM calls as functions
-
Builder Code by Builder.io (formerly Fusion) AI agent turning Figma frames into design-system-matched pull requests
-
Chroma Simple open-source vector and full-text search, from laptop to cloud
-
Cline Open-source coding agent for your editor, terminal and desktop
-
Codacy Code-quality and security gates for pull requests, plus guardrails for AI code
-
CodeGPT Model-agnostic AI coding agent for VS Code, with your keys or local models
-
CodeParrot Hugging Face's 2021 teaching model for training a code LLM from scratch
-
CodeRabbit AI pull-request review, now with PR triage and security monitoring
-
CoderPad Live coding interviews where candidates can use AI and you can watch how
-
Codiga Acquired by Datadog in 2023 — standalone code analysis shut down
-
Comet ML Open-source Opik for LLM apps, plus classic ML experiment tracking
-
Continue.dev Open-source AI coding assistant — discontinued, acquired by Cursor
-
Databutton Pivoted to Riff: ERP-connected AI agents for procurement teams
-
Factory Droid coding agents across the software lifecycle, from CLI to on-prem
-
Gemini Enterprise Agent Platform (formerly Vertex AI) Google Cloud's AI and agent platform, renamed from Vertex AI in 2026
-
GPT Pilot Pioneering open-source AI dev agent, now unmaintained
-
Graphite Stacked diffs, a merge queue, and an AI reviewer for GitHub
-
Groq Fast open-weight model inference on Groq's LPU chips, now beside Nvidia GPUs
-
InCoder Open-source code infilling model — generates code in the middle of existing functions, not just at the end.
-
JetBrains AI AI Assistant and coding agents built into JetBrains IDEs
-
Marblism Seven subscription AI "employees" for inbox, social, leads, web, SEO, calls
-
Maxim AI Agent simulation, evaluation and tracing, plus the open-source Bifrost gateway
-
Mentat Terminal-based multi-file coding assistant
-
Milvus Open-source vector database for billion-scale search, now lake-native
-
Momentic AI-native E2E, visual, API, and accessibility testing — write tests in plain English, ship with confidence.
-
NeuBird AI Agentic AI SRE that autonomously prevents, investigates, and resolves production incidents before engineers are paged.
-
Opik Open-source LLM and agent tracing, evals and optimisation you can self-host
-
Pieces On-device work memory with AI recall, summaries and an MCP server
-
PolyCoder CMU's 2022 code LLM — an early open, multi-language code model
-
PromptLayer A prompt CMS with evals and observability for AI teams
-
PyCodeGPT A small, archived 2022 Python code model from Microsoft research
-
Qodo Agentic AI code review and rules governance for engineering teams
-
Sherlocks AI AI SRE that investigates alerts and posts root cause in Slack or Teams
-
Smol Developer Junior-engineer AI for spec-to-code
-
Softgen Prompt-built web apps on managed Supabase, with your code on GitHub
-
Sourcery AI pull-request reviewer with security scanning, free for open source
-
StackBlitz Browser-based IDE on WebContainers, from the company behind Bolt.new
-
Stainless Acquired by Anthropic; SDK generator closed to new signups (2026).
-
Tabby Self-hosted, open-core alternative to GitHub Copilot
-
Thunders Plain-language AI agents that write, run and repair end-to-end tests
-
Tinybird Managed ClickHouse that turns SQL queries into production APIs
-
ToolJet Open-source low-code builder for internal tools and AI apps
-
Unstructured Open-source and hosted document parsing for LLM-ready data
-
Weaviate Open-source vector database with hybrid search built in
-
Zencoder Spec-first agent orchestration across frontier models, plus IDE agents