Build with AI

Latitude gives you full visibility into every Claude Code session — the system prompt it uses, every tool call it makes, the bash output, and the files it touches. Install one npx command and you get traces across your terminal, Claude Desktop, and your IDE. Use it to catch failures, track cost per session, and understand exactly what your AI coding agent is doing.

General Compute is an AI inference provider built on purpose-built ASICs — not GPUs — delivering over 1,000 tokens per second with a time-to-first-token under 300ms. You get a fully OpenAI-compatible REST API, so it works as a drop-in replacement for existing integrations, plus $100 in free credits to get started. It also offers custom dedicated deployments and bring-your-own-model options for larger workloads.

deepsec is an open-source security harness that uses AI coding agents to find vulnerabilities in your codebase. You run it on your own infrastructure — locally or scaled across Vercel Sandboxes — and it uses your existing Claude or Codex subscription to scan, investigate, revalidate, and export actionable findings with severity ratings. Vercel built it to scan their own monorepos and now open-sources it for any team to use.

Kdeps lets you build AI agent workflows in YAML and run them anywhere — locally with Ollama or llamafile (no API key needed), or in the cloud with OpenAI, Anthropic, or Groq. You write one workflow file, and Kdeps turns it into a running REST API, bot, or multi-agent pipeline. When you're ready, export it as Docker, Kubernetes, a single binary, or a bootable ISO — same file, no rewrites.

Kumbukum is an open-source, team-shared memory layer that connects your notes, decisions, emails, links, and Git context to Claude, Cursor, ChatGPT, and any MCP-compatible tool. Instead of re-explaining your project background at the start of every AI session, your team's full knowledge library is already there. It includes an Email Command Center that reads incoming email against your shared library and drafts context-grounded replies automatically.

AgenticCalling gives your AI agent the power to make real phone calls. You connect via MCP (Claude, ChatGPT, Cursor) or a REST API, and the service handles the full call: navigating phone menus (IVRs), leaving voicemails, and returning a structured JSON transcript when it's done. At $0.09 per minute flat — with no commitment on the pay-as-you-go plan — it runs up to 70% cheaper than ElevenLabs and about 30% cheaper than Vapi.

Tollecode is a local-first AI coding assistant that lets you delegate real engineering tasks — file reads, shell commands, and multi-step sub-agents — to an AI that runs on your machine. You get both a desktop app and a CLI, and you choose your own AI provider: Claude, OpenAI, Gemini, Ollama, or any OpenAI-compatible endpoint.

Glide builds custom business apps directly from your spreadsheets and databases — no coding required. Turn a Google Sheet, Excel, or Airtable into a working mobile or web app in minutes. Free for testing. Maker plan at $25/month.

Replit Agent is best for small business owners and non-technical founders who want to build real web or mobile apps without hiring a developer. You describe what you want in plain language, and the AI writes and deploys the code for you. It handles the full stack — database, hosting, authentication — so you don't have to set anything up. If you have a simple app idea, a client portal, or an internal tool you've been putting off, this can get you there fast.

Codeium gives developers AI-powered code suggestions and a chat assistant inside their existing code editor, supporting over 70 languages with a free individual plan.