10 AI Engineering Tools I Keep Pinned While Building
(1) models.dev
Context windows, pricing and rate limits for every model in one queryable database. Faster than opening five pricing pages.
(2) artificialanalysis.ai
Speed, cost and quality on the same axes. The latency-vs-price scatter is the one chart worth checking before you commit.
(3) build.nvidia.com/models
90+ open models running in the browser. You can shape a prompt against a 70B before you rent a single GPU hour.
(4) context7.com
Pipes current library docs into your coding agent, so it stops calling functions that were deprecated eight months ago.
(5) deepwiki.com
Points at a GitHub repo and turns it into something you can interrogate. Good for the hour before you decide to adopt a dependency.
(6) gitingest.com
Flattens a public repo into one file you paste straight into a long-context model. Pairs with the above.
(7) smithery.ai/servers
7,400+ MCP servers, searchable. The registry that tells you whether the integration you need already exists.
(8) epoch.ai
Compute, training cost and scaling trends, with the underlying data published. It is the one place I will cite a number from without checking it twice.
(9) promptingguide.ai
Prompting techniques mapped back to the papers that introduced them. Useful mainly because it shows you what has already been tried.
(10) roadmap.sh
The AI engineering path, laid out as a dependency graph. Most useful for spotting the thing you skipped.
The tooling layer got good faster than the models did, and almost nobody is pricing that in.
What is the tab I am missing?
I keep ten tabs pinned while I build. None of them are ChatGPT.
Here are the AI engineering tools I keep coming back to:
1. models.dev
A queryable database of AI model specifications, pricing, capabilities, context limits and other metadata. Useful when comparing models before checking the provider's official documentation.
2. Artificial Analysis
Compare AI models across intelligence, price, output speed, latency, context window and other performance metrics. The cost-vs-performance comparisons are particularly useful when choosing a model.
3. NVIDIA Build
A browser-accessible catalog for trying and deploying NVIDIA NIM models, including free endpoints and downloadable models.
4. Context7
Pulls current, version-specific documentation into AI coding workflows, helping reduce outdated API calls and hallucinated library usage.
5. DeepWiki
Give it a repository and explore AI-generated documentation and explanations of the codebase. Useful when evaluating an unfamiliar dependency or understanding a large project.
6. Gitingest
Turns a Git repository into a prompt-friendly text digest of the codebase, making it easier to provide repository context to an LLM.
7. Smithery
A searchable registry for MCP servers. Before building an integration from scratch, it's worth checking whether an MCP server already exists for the service you need.
8. Epoch AI
A strong source for understanding AI scaling, training compute, model trends and the economics behind increasingly large AI systems. Its public database tracks thousands of machine-learning models.
9. Prompt Engineering Guide
Useful as a reference for prompting techniques and the research behind different approaches.
10. roadmap.sh
The AI Engineer roadmap gives you a structured view of the skills and technologies that make up modern AI engineering.
The tooling layer around AI has become surprisingly deep.
Model selection, documentation retrieval, repository understanding, MCP discovery, evaluation and engineering roadmaps are increasingly becoming part of the actual development workflow.
But I think one important category is still missing:
AI observability and evaluation.
Once an AI application reaches production, knowing which model to use is only half the problem.
You also need to know:
Why did the model produce this answer?
How often are responses failing?
Which prompts are degrading?
How much is each request costing?
Are latency and quality changing over time?
Did a model update make the application worse?
That's the tab I'd add next.
What AI engineering tool would you add to this list?