Top Development & Code AI Tools

Open-source multilingual speech recognition model from OpenAI.

OpenAI Whisper is an open-source automatic speech recognition model that transcribes audio to text in over 90 languages, with built-in translation support. Trained on a large, diverse audio dataset, it maintains accuracy across accents, background noise, and technical vocabulary, and can be freely integrated into developer applications.

• Open-source, free to self-host
• Multilingual transcription across 90+ languages
• Built-in translation support
• Fine-tunable for custom use cases

Secure cloud sandboxes for running AI-generated code safely.

E2B provides isolated cloud sandboxes that AI agents and code-interpreter features can execute untrusted, AI-generated code inside — spinning up in seconds, with filesystem and internet access, so developers don't have to run model output directly on their own infrastructure.

• Fast-starting isolated code sandboxes
• Built for AI agent code execution
• SDKs for Python and JavaScript

AI-ready web scraping API that turns any website into clean, structured data.

Firecrawl crawls websites and converts pages into clean markdown or structured JSON specifically formatted for feeding into LLMs, handling JavaScript rendering, pagination, and rate limits so developers don't have to build a scraper for every AI data-ingestion pipeline.

• LLM-ready markdown/JSON output
• Handles JavaScript-rendered pages
• Full-site crawling and single-page scraping

AI-native search API built for retrieval by LLMs and AI agents.

Exa (formerly Metaphor Systems) is a search API built specifically for AI applications — it uses embeddings-based search to find semantically relevant web pages for a query, rather than keyword matching, making it a common backend for AI research agents and RAG pipelines.

• Embeddings-based semantic web search
• Built for LLM/agent retrieval, not humans
• Content and similarity search endpoints

AI-driven tool for finding and prioritizing risky open-source dependencies.

Endor Labs analyzes open-source packages a codebase depends on, using AI-driven risk scoring to flag which ones are actually reachable, risky, or worth patching first.

• AI-prioritized dependency risk scoring
• Reachability analysis
• SBOM and supply-chain visibility

AI-assisted code review and stacked pull-request workflow tool.

Graphite speeds up code review with AI-generated PR summaries and suggested comments, alongside a stacked-diff workflow that helps engineering teams ship smaller, faster-reviewed changes.

• AI-generated PR summaries
• Stacked pull-request workflow
• Merge queue and review analytics

Enterprise LLM provider behind the Jamba model family, built for long-context tasks.

AI21 Labs is an Israeli AI company building the Jamba family of large language models on a hybrid transformer/state-space architecture optimized for long context windows. Its API and AI21 Studio platform target enterprise use cases like document summarization, knowledge management, and customer service automation in regulated industries.

• Jamba hybrid transformer/state-space models
• Optimized for very long context windows
• Enterprise focus with regulated-industry customers
• Developer API via AI21 Studio

Developer-focused speech-to-text API for real-time and batch transcription.

Deepgram provides a speech-to-text and voice AI API built for developers, supporting real-time and batch audio transcription across many languages and accents. Features include speaker diarization, sentiment analysis, and custom vocabulary training, aimed at call center analytics, voice assistants, and media captioning at scale.

• Real-time and batch transcription
• Speaker diarization and sentiment analysis
• Custom vocabulary training
• Low-latency, high-scale API

AI coding assistant built into IntelliJ IDEA, PyCharm, WebStorm, and other JetBrains IDEs.

JetBrains AI Assistant is an AI coding assistant integrated directly into JetBrains IDEs, offering code completion, code explanation, refactoring suggestions, and unit test generation with deep awareness of the surrounding project context. It also generates commit messages and helps with code review.

• Deep project-context-aware code completion
• Refactoring and unit test generation
• Commit message generation
• Native integration across JetBrains IDEs

AI code quality platform for automated test generation and pull request review.

Qodo (formerly CodiumAI) is an AI code quality platform focused on generating reliable unit and integration tests and reviewing code for correctness ('code integrity'). Its Qodo Merge feature automatically reviews pull requests, flagging bugs, security issues, and style problems, integrating with GitHub, GitLab, and popular IDEs.

• Automated unit and integration test generation
• AI pull request review (Qodo Merge)
• Flags bugs, security issues, and style problems
• GitHub, GitLab, and IDE integrations

Managed vector database for building AI search and RAG applications.

Pinecone is a fully managed vector database used to store embeddings and run fast similarity search at scale — the backbone of most retrieval-augmented-generation (RAG) AI applications, semantic search products, and recommendation engines.

• Fast vector similarity search at scale
• Fully managed, no infra to run
• Metadata filtering and hybrid search

Open-source vector database for AI search and RAG applications.

Weaviate is an open-source vector database that stores data objects alongside their vector embeddings, letting developers combine semantic search with structured filters. It can be self-hosted or used as a managed cloud service, and integrates with most popular embedding models.

• Open-source and self-hostable
• Hybrid semantic + keyword search
• Modules for popular embedding providers

Data framework for connecting LLMs to your own documents and data.

LlamaIndex is an open-source data framework that helps developers load, index, and query private or domain-specific data with large language models — the connective layer commonly used to build retrieval-augmented-generation (RAG) pipelines and AI knowledge agents.

• Data connectors for 100+ sources
• RAG pipeline building blocks
• Agent and workflow orchestration

AI-powered search and discovery API for websites and apps.

Algolia provides a fast, typo-tolerant search API that developers drop into e-commerce and content sites, now extended with AI features like natural-language query understanding, personalized ranking, and generative answers over a site's own content.

• Millisecond-fast, typo-tolerant search
• AI-powered relevance and personalization
• Generative answers over site content

AI-assisted localization management platform for apps and websites.

Lokalise is a translation management system built for product teams — it centralizes strings from apps, websites, and docs, uses AI to pre-translate and suggest terminology, and plugs directly into developer workflows via CLI and API.

• AI pre-translation and suggestions
• Developer-friendly CLI/API/CI integrations
• In-context screenshot-based translation editor

AI-powered digital accessibility testing platform for websites and apps.

Evinced scans websites and mobile apps for accessibility issues using AI-driven analysis that goes beyond static rule-checking, prioritizing the fixes with the most real-world impact and integrating into engineering workflows for ongoing compliance.

• AI-driven accessibility issue detection
• Prioritized fix recommendations by impact
• Integrates into CI/CD and engineering workflows

AI-powered web accessibility widget and compliance platform.

UserWay adds an AI-driven accessibility widget to websites that adjusts contrast, text size, and navigation for users with disabilities, alongside automated scanning tools that help site owners work toward WCAG and ADA compliance.

• One-line install AI accessibility widget
• Automated WCAG/ADA compliance scanning
• Screen-reader and keyboard-navigation support

Web search API designed for AI agents and RAG applications.

Tavily is a search API built specifically for LLM agents to answer questions — it returns pre-summarized, source-cited results optimized for feeding directly into a model's context window, rather than raw search-engine results pages meant for humans.

• LLM-optimized, pre-summarized search results
• Source citations included in results
• Built for real-time agent research tasks

Framework from LangChain for building stateful, multi-step AI agents.

LangGraph is a library from the LangChain team for building AI agents as explicit graphs of steps rather than a single prompt-response loop, giving developers control over branching logic, retries, and long-running multi-agent workflows.

• Graph-based agent orchestration
• Built-in state persistence and retries
• Works with any LLM provider

Serverless cloud platform for running AI/ML workloads and GPUs on demand.

Modal lets developers run Python functions in the cloud — including GPU-backed AI inference and training jobs — without managing servers, scaling automatically from zero and billing by the second, popular for deploying custom AI models and batch jobs.

• On-demand serverless GPUs
• Scales to zero, pay-per-second billing
• Simple Python-native deployment