Build & Connect

Helicone monitors, debugs, and improves your AI applications. Tracks every LLM API call so you spot errors, control costs, and understand performance. Works with 100+ AI models through a single integration. Free: 10,000 requests/month. Pro: $79/month.

Google's AI assistant — duplicate of Gemini entry. See that listing for full review.

An AI executive assistant that manages your inbox, calendar, meetings, and follow-ups over text. You can also build custom AI agents that work across your apps.

Open-source framework for lightweight desktop and mobile apps using web technologies. Uses OS native renderer instead of bundling Chromium — apps as small as 600KB vs Electron's 100MB+. Rust backend for security. Completely free and open source. MIT license.

Run AI models on your own computer — free and private. Download 100+ open-source models (Llama, Mistral, Qwen) and run locally. OpenAI-compatible API works with existing tools. Completely free. Open source. Mac, Windows, Linux.

GitBook is a technical documentation platform that helps teams write, publish, and maintain product docs that connect with users and AI agents. It supports docs-as-code via Git Sync, includes an embedded AI Assistant that answers user questions, and ships every site with an MCP server for AI agents.

MindFort helps you find and fix security vulnerabilities in your software before attackers do. You point AI-powered agents at your app or API, and they test it like a real attacker would — then deliver code patches (pull requests) you can merge in minutes. You get results in under an hour with less than 0.1% false positives.

JetBrains makes professional IDEs — IntelliJ IDEA (Java), PyCharm (Python), WebStorm (JavaScript), PhpStorm (PHP). Intelligent code completion, real-time error detection, and refactoring tools. All Products Pack bundles everything for $249/year per developer. Includes AI assistant Junie, TeamCity for CI/CD, and YouTrack for issue tracking. Students and educators get free licenses.

AssemblyAI gives your business a way to turn speech into text using a simple API. If you record customer calls, run a podcast, or capture meeting audio, AssemblyAI can transcribe it accurately in 99 languages — and then go further by identifying speakers, detecting topics, summarizing conversations, and flagging sensitive information automatically. Beyond basic transcription, the platform includes Audio Intelligence features like sentiment analysis, content moderation, and chapter detection. Their LeMUR framework lets you connect transcripts directly to large language models, so you can pull insights, generate action items, or answer questions about your audio without building a custom pipeline. AssemblyAI is built for developers. You integrate it through a REST API or official SDKs, and you can also connect it to no-code platforms like Make.com. Pricing is pay-as-you-go starting at $0.15 per hour of audio, with a $50 free credit to get started.

Ultrareview helps your dev team catch real bugs before they merge into your codebase. You run one command and a fleet of AI reviewer agents checks your code in a remote cloud sandbox — no slowdown on your local machine. Every bug it reports is independently verified, so you spend less time chasing false alarms.