The argument in one line.
Using ref.tools and Exa AI as MCP servers reduces documentation token consumption by 97%, enabling Claude Code to complete Tailwind refactoring audits in 2,800 tokens versus 98,000 in Cursor by fetching only the precise context each task requires.
Read if. Skip if.
READ IF YOU ARE…
- You're a full-stack engineer shipping features in Claude Code or Cursor who's frustrated with token bloat from dumping entire documentation into prompts.
- An AI engineer building agentic workflows who needs multi-step tool calling without exhausting context windows on documentation retrieval.
- A developer new to MCP servers who wants a concrete walkthrough of installing ref.tools and Exa AI across Claude Code, Cursor, Codex, and Factory Droid.
SKIP IF…
- You're already using MCPs effectively in production and looking for advanced optimization techniques beyond the setup-and-benchmark scope here.
- You work primarily in non-coding domains or don't use Claude Code, Cursor, or similar AI coding environments regularly.
The full version, fast.
Two MCP servers, ref.tools and Exa AI, dramatically cut the token bloat that wrecks AI coding sessions by fetching only the documentation the model needs instead of dumping entire reference sets into context. Install them in Claude Code, Cursor, Codex, or Factory Droid by pasting the provided command with your API key, then explicitly invoke them in prompts ("use ref MCP and exa MCP") while working in plan or spec mode so research happens before any code is written. On a live Tailwind v4 refactor, this approach consumed 2,800 tokens in Claude Code versus 98,000 in Cursor doing the same audit, a thirty-five-fold efficiency gain. Pair targeted retrieval with cheaper agentic models like Haiku for research-heavy refactors to ship faster without exhausting the context window.
Free for members
Chat with this breakdown — free.
Sign in and you get 23 free chat messages on us — ask for the hook, quote a framework, find the exact transcript moment, generate a markdown action plan. Bring your own key when you want unlimited.
Create a free account →Where the time goes.
00:00 – 02:05
01 · The context rot problem
Ray introduces the concept of context rot — LLMs getting dumb as the window fills with irrelevant documentation. Establishes why targeted doc-fetching MCPs beat brute-force token dumps.
02:05 – 04:37
02 · Installing ref.tools in Claude Code
One-command install via `claude mcp add`, API key walkthrough, and security warning about committing keys to public repos.
04:37 – 05:38
03 · Installing Exa AI in Claude Code
Same pattern as ref.tools. API key generation and paste into terminal. Both MCPs now available globally across any project.
05:38 – 12:43
04 · Live demo: Tailwind v4 refactor on Anime Leak
Haiku 4.5 runs a full codebase audit using both MCPs. Ray watches Claude Code research documentation, build a phased implementation plan, and hit /context — revealing only 2,800 tokens used.
12:43 – 15:00
05 · The token comparison: Claude Code vs Cursor
Cursor's plan mode finished the same task using 98,000 tokens. Side-by-side makes the 35x gap visceral. Ray notes Cursor asks clarifying questions; Claude Code doesn't.
15:00 – 16:50
06 · Setting up MCPs in Codex
Codex uses config.toml instead of JSON. Ray edits the file in Cursor, pastes the API key, verifies with `codex mcp list`.
16:50 – 17:50
07 · Setting up MCPs in Factory Droid
Copy the MCP JSON block from Cursor's tools panel directly into .factory/mcp.json. Same pattern, different file path.
17:50 – 18:51
08 · Pro tip + CTA
Always call MCPs explicitly in your prompt. Use plan mode (shift-tab) before writing code. Closes with a pitch for his 1337 coaching intensive and a forward-looking take on agentic AI.
Lines worth screenshotting.
- The same Tailwind v4 refactor used 2,800 tokens in Claude Code with ref.tools versus 98,000 tokens in Cursor without it — a 35x difference from targeted documentation fetching.
- Context rot — the model getting progressively dumber as documentation floods the context window — is the root problem that ref.tools and Exa AI are specifically designed to solve.
- ref.tools indexes public documentation, PDFs, GitHub repos, and private documentation, then serves only the specific section the agent needs for the current step rather than dumping everything.
- Exa AI provides high-quality fast search optimized specifically for coding tasks — combining it with ref.tools gives the agent both targeted doc retrieval and broad technical search.
- Adding 'use context seven' or 'use ref' to a prompt is enough to activate the MCP — the agent automatically calls the server to fetch only what is needed for that specific operation.
- API keys for MCP servers should never be committed to a public repo — anyone who clones the repo can use your key and drain your credits.
- Installing a global MCP server from the root user profile makes it available to every new project automatically without repeating the setup command per project.
- Models operate best with just the right information for just the right task — overloading the context with all available documentation before any task begins is the opposite of optimal.
- Feeding 100,000 tokens of documentation into a context before coding starts was common practice before targeted MCPs — it was slow, expensive, and degraded model output quality.
- The combination of ref.tools for documentation and Exa for search makes Claude Code feel like having multiple senior developers who have already read all the relevant docs before writing a line.
- Plan mode plus these two MCPs allows building features that are near-complete before the coding phase begins — the agent arrives at implementation with full context rather than guessing.
- Documentation currency matters: libraries update constantly, and an MCP that fetches live docs eliminates the training-data cutoff problem that causes AI to generate deprecated API calls.
Targeted Documentation Fetching via MCP Uses 35x Fewer Tokens Than Brute-Force Doc Dumps
Claude Code setup
Ray Fernando's live benchmark shows that context rot — the model getting progressively dumber as irrelevant documentation fills the window — is the primary source of AI coding quality degradation, and that two MCPs which fetch only the relevant sections solve it at 2,800 tokens versus 98,000.
01The context rot problem
- Context rot: the model gets dumb as the window fills with irrelevant documentation — the failure is not the model, it is the information management strategy
- The fix is fetching only the relevant sections for the specific task, not the entire documentation set
02Installing ref.tools in Claude Code
- One-command install via claude mcp add — API key generation and paste is the entire setup
- Security warning: do not commit API keys to public repos — global install keeps the key in Claude's config, not the project folder
04Live demo: Tailwind v4 refactor
- Haiku 4.5 ran a full codebase audit with both MCPs — the model researched documentation, built a phased plan, and completed the task at 2,800 tokens
- The /context check after the task is the verification step — always confirm token usage against your expectations before starting expensive downstream work
05The token comparison: Claude Code vs Cursor
- Same task, same libraries: Claude Code at 2,800 tokens, Cursor at 98,000 — a 35x difference driven entirely by how documentation was fetched
- Cursor asked clarifying questions; Claude Code did not — the behavioral difference is a secondary finding worth tracking across projects
08Pro tip + CTA
- Always call MCPs explicitly in your prompt — autonomous MCP selection by the agent produces less consistent routing than explicit direction
- Use plan mode (shift-tab) before writing code — surface the documentation needs before implementation, not during it
Terms worth knowing.
- MCP server
- A Model Context Protocol server — a standardized integration that exposes specific tools or data sources to an AI coding agent, allowing the agent to call those tools during a task rather than relying solely on its training data.
- context window
- The total amount of text an AI model can hold in memory during a single session — measured in tokens — determining how much code, documentation, and conversation history the model can consider simultaneously before older content is dropped.
- context rot
- The degradation in an AI model's response quality that occurs when the context window fills with large, loosely relevant content — such as entire documentation dumps — leaving the model without space for precise, task-specific reasoning.
- ref.tools
- An MCP server that indexes public and private documentation — including PDFs, GitHub repos, and websites — and retrieves only the specific excerpts an AI agent needs for a given query, instead of dumping entire documentation files into the context window.
- Exa AI
- A search API and MCP server optimized for code-related queries that returns semantically relevant technical results — used as a complement to documentation-indexing tools to give AI coding agents fast access to up-to-date information during agentic tasks.
- agentic search
- A search pattern where an AI agent issues multiple iterative queries across tools and documents to build up the context needed for a task — as opposed to a single one-shot lookup — designed to gather just enough relevant information without flooding the context window.
- tool call
- An action taken by an AI agent to invoke an external function or API — such as reading a file, querying a database, or calling an MCP server — logged as a discrete step during agentic task execution and consuming tokens from the context window.
- plan mode (Claude Code)
- A Claude Code operating mode activated by pressing Shift+Tab that causes the agent to analyze the codebase and outline an implementation plan without writing any code — allowing the developer to review and approve the approach before execution begins.
- Tailwind v4
- Version 4 of the Tailwind CSS utility framework — a significant rewrite that replaces the JavaScript configuration file with CSS-native design tokens and variables, changing how custom design systems are structured and maintained in a project.
- design token
- A named variable that stores a single design decision — such as a color, spacing value, or font size — defined once in a central stylesheet and referenced throughout a codebase to ensure visual consistency and make global style changes manageable.
- context pruning
- The automatic process by which an AI model discards older or lower-priority content from its context window when that window becomes full — a risk in long agentic sessions where relevant information gathered early in the task may be dropped before it is needed.
- OpenAI Codex (agent)
- An AI coding agent from OpenAI that can execute multi-step software development tasks autonomously in a sandboxed environment — capable of reading codebases, running tests, and committing code changes.
- Factory Droid
- An AI coding agent platform (by Factory) that automates software engineering tasks — configurable with MCP servers so the agent can access external documentation and tools during autonomous code generation and review workflows.
- config.toml
- A plain-text configuration file written in the TOML format — used by tools like Codex CLI to store settings, including which MCP servers are enabled, in a structured, human-readable key-value format.
Things they pointed at.
05:38productAnime Leak app
15:00toolCodex CLI
16:50toolFactory Droid
17:50product1337 Intensive coaching offer
Lines you could clip.
00:34
“I used to use up almost 100k tokens just feeding in tons of documentation.”
00:53
“LLMs actually operate best when they have just the right information for just the right specific task.”
12:00
“Our MCP servers and tool calls only use a total of 2,800 tokens — which is only 1.4% of the context window.”
02:05
“It's almost like as if I hired several developers to read the documentation and implement the code for us.”
16:50
“Make sure you use plan mode so that it gathers all that specific context before you start writing code.”
Word for word.
Read-along
Don't just watch it. Burn it in.
See every word as it's spoken — crank it to 2× and still catch all of it. The same dual-channel trick behind Amazon's Kindle + Audible.
metaphoranalogy
00:00So I'm gonna show you how to get two of my favorite MCP set up. And the two are ref dot tools and the other one's called Exa AI. And the reason why I like ref dot tools compared to every other MCP that you could use to search for documentation is .
00:15So that's when the LLMs start getting really dumb as soon as they try to help you out with information. And if you don't know, whenever you're using a language model, there's a cutoff date for all the data that they have trained so far. And so that's why I choose these two different MCP servers because they're the most efficient.
00:31If you've ever used any other MCP server, all of a sudden, you're gonna start slamming in a whole bunch of tokens from documentation. I used to do that before.
00:39I used to use up almost a 100 k tokens just feeding in tons of documentation for the new AI SDK, my authentication, my database provider, all these extra rules.
00:49But the LLMs actually operate the best when they have just the right information for just the right specific task.
00:57And so in comes ref dot tools. So this actually provides the actual context that's needed for these agents. And the way that it works is it'll grab public documentation and any private documentation that you also give it as well.
01:11PDFs, GitHub repos, any other sites. It will create them and indexes and actually make them available in Cloud Code, in Droid, in Codex, in Cursor, and those are the four that we're gonna go ahead and cover today as far as trying to get this installed for us to make it work.
01:27A lot of times when you're doing tool calling, what ends up happening is that there's gonna be multiple requests and multiple steps that happen along the way. And ref dot tools is kinda made for this agentic search so it doesn't kill your context window.
01:40So this combined with the other one called XAI, this actually has a really high quality fast search for specifically coding tasks. And if you combine these two together, you're gonna be cooking.
01:53And I have shipped features starting in plan mode using these two MCP servers to help gather the right context. And by the time it's writing the code, it's almost like as if I hired several developers to read the documentation and and implement the code for us.
02:07How do we go ahead and configure them? So in ref dot tools, they actually make it really easy for you. All you do is just go ahead and once you've already created an account, uh, you just go to MCP and in Cloud Code, all we're gonna do is install it via command line.
02:20So this API key here is what you're gonna actually put into this command. So once you hit copy, we're gonna open the terminal now. And then I'm just gonna go ahead and just paste this in here.
02:30Now it says added the MCP server with ref and has a local config headers of this. And so this means that the actual API keys here, and I think it's gonna be able to do that. So now if I run Claude and I run slash m c p, it's gonna try to connect.
02:46Now we actually have the ref dot tools actually connected for us. And we have to like quit an existing session like we did and try to log in again. And so you'll see it available once you do cloud mcps, uh, space list.
02:57So because it's in my root user, it's gonna be available to any new project I start. So don't be afraid if you add this in, you don't have to keep adding this into every project.
03:05You just have to add it in once and once you add it in here, you're pretty much good to go. By the way, if you notice, we actually have the API key currently in the MCP server. You typically don't wanna check those types of things into your project as code because that's gonna be available to everyone.
03:20So if you have the repo that's public, anyone who clones that can use your specific key and eat up all your credits as an FYI. So the next one we wanna go ahead and get configured is called exa.
03:31The command they give us is claud m c p add exa and the same with our API key. So what we're gonna go ahead and do is hit copy here. Then I'm gonna go back to cloud code in the terminal here.
03:41And what we wanna do is go ahead and place this thing called here, this is your API key. And so in this section here, what we do is just gonna go ahead and delete. And then I'm gonna grab the API key from the following API dashboard.
03:52If you've not generated a key, you can go into here and create a new key. So we can create one called, Claude code YOLOBRO.
04:03And let's go ahead and create the key. And then I'm gonna go ahead and copy this one. And then I'm gonna go into my terminal now.
04:09And then I'm just gonna go ahead and paste that key there. I'm just gonna go ahead and hit enter. It's gonna go ahead.
04:14It's now added the MCP server with the following command to the local config. And so now it's actually the file that's modified is in my local folder here. So you may notice that it could take a while for it to connect, and you can always just do Claude MCP list.
04:31And that'll basically show you what's actually currently active and available as far as MCP servers.
04:37So Haiku has just come out. It's the 4.5 model supposed to act really fast. And why don't we just try using the the latest model for a task that I'm trying to do?
04:45And in this case, we're basically just gonna ask the model to see if you can use the latest tailwind before and search through our code base. My code base isn't really fully utilizing the power of tailwind before.
04:58And I want you to take a look at my code at all the different places because I think I have a lot of stuff that I've hard coded in for tail and v three and I need it to be all unified.
05:08So we're taking advantage of the new tail and v four system. I want you to use ref and use exit MCPs first to read through the documentation to understand how this type of system works, as well as reading through all the different parts of my code to understand the current design system.
05:26And we should be using Tailwind v four along with chat c n. So right now we're basically having Claude code look through my entire app. And if you don't know what my app currently does, it's called anime leak.
05:39It allows you to upload an image, any real image of the real world. And you'll actually start to see like basically anime start leaking into the photo, which is kind of fun.
05:48So you can see me right here with my sandwich got replaced with this nice hand drawn kind of really whimsical type of thing. So in here you can kind of see all the different generations I've had, and then it lets you like share them with friends and so forth. In my original actual design, everything was in light mode.
06:04And since I forked this project from my friend Mickey, who currently works over at Convex, it was in this specific color theme.
06:12So the landing page as you can see is kinda a little bit more dark theme and it has these different types of things that are going on. And we wanna have more of like a consistency for design hierarchy and everything here. So right now, Cloud Code is actually searching through and it's gonna try to understand the code base fully.
06:27Now it's researching Tailwind Design Systems. And I don't know if it's actually fully utilizing the MCP's here. So now it's actually saying, yes, I want to use these different tools now to use the ref documentation.
06:38So the query is gonna do is Tailwind v four CSS variables design tokens customization, and yes, and don't ask again for this MCP.
06:46And the reason why I'm thinking Haiku is the right thing for this type of task is one, it's a lot cheaper. And for this type of task, I don't need a lot of intelligence. I just need a lot of work to be done.
06:56And so by handing this off to an agentic model like this, it's it's gonna be able to gather a lot of research in one side of it, but also look through the code and and look for specific patterns of things which is what I want to do. And then the utilizing of these MCP's will keep the documentation very light, but grab enough information that it needs to know to do this type of pattern matching which is gonna be really awesome here.
07:16So so while that's going, what I'm gonna go ahead and do is pull this up in cursor and kinda show you how I configured those there. We're gonna go to this little gear here and they have actually dedicated tools and MCPs. If you scroll down, these are all the MCP servers I have connected.
07:30You can just hit add custom MCP server and this is a JSON file that actually has everything configured here. And what you wanna go ahead and do is you'll go to the the website where it says use ref they're gonna give you this exact thing to paste into cursor.
07:43Add the cursor in one click. And if we go into here, it's actually gonna say, do you want to install the MCP servers? And you see ref mentioned here with the following URL and the API key.
07:53So this already has the API baked in and you hit install. Once you hit install, it's gonna actually show up just like it does here.
08:00Pretty simple. It's really nice. So we're gonna basically do the same thing with Exa.
08:03If we go back to Exa, and we go to the dashboard, and we go to Exa MCP.
08:10So the values that we wanna grab is gonna be this one here. See Exa, and then has this little bracket here and this bracket here.
08:17This is really really important because at the higher level which is MCP servers, that's your collection of all of your MCP servers. So we just get this value just like that.
08:26And you see that there's an additional thing that's added to my specific MCP server which is this thing here. It's the Exa API key equals and this is the part where you wanna grab that API key from Exa and then put it into this last section here.
08:42And so when you open your project, you just make sure to toggle on and this is green and that means you're pretty much good to go. And I'm gonna give the agent the same type of task to use the same MCP service as well. I'm just gonna copy this exact actual prompt and put this inside the cursor, so that we can kinda see them both kinda cook right here.
08:57So so now, what we can do as well is we can put the agent into plan mode. So what happens in plan mode is no code gets written, but then we'll have all this analysis done and we'll have MCP's be able to be utilized inside of cursor. And so cursor is gonna do all the work for us, which is really nice.
09:14If I do command e, this is gonna pop up a dedicated specific window for our agents, so that we can do this type of planning.
09:22Now, we get a straight up dedicated window to watch the agent really work. And to me, this is like a really nice way to work as you can kinda see it's a little more spread out.
09:31And as we are gonna start modifying files or working with anything in the code base, we have what I feel like is a much more elegant view, all self contained in in one type of thing. So to get to this agent window, you have to hit command e.
09:44You And may have to enable it in the settings as well. So as you can kinda see it's narrowing down the scope as it discovers information to make sure that everything is the latest information is up to date. And so this is a lot more efficient than web search after web search after web search.
09:59Sometimes they'll run a parallel tool call to do a lot of web search and then you'll start to flood the context window and a lot of times these, you know, agents will have to start pruning information. And sometimes you can have really relevant information pruned out.
10:12And that's why I prefer these two MCP's. One of my favorite things about cursor is how fast it is at gathering contacts and information throughout my file system. Because if we go back to Cloud Code right now, while it's still running in the terminal, it is still running all these different tasks.
10:27We're still on the research side and you can see all the different tool calls it's been doing to get all the relevant information that it needs for this task. And it takes a little while, right? So it's now auditing current code base for our code values, create a comprehensive design token system and global dot CSS.
10:42But this is actually kind of already giving us a plan here. So I have a comprehensive understand. Let me update the to do list and create a detailed implementation plan.
10:49So it actually created a plan for us without me actually asking for it, which is really nice. Right? So this is my current state.
10:55My global CSS is already using Tailwind v four correctly. However, there are opportunities to expand and unify the system. But as you vibe code stuff, the agents are gonna prefer what's in their training set, so they're gonna dump in a bunch of v three type of systems, but here's the gaps to address.
11:10So there's like missing semantic tokens, means hard coded values are scattered, magic numbers and components, and all this kind of stuff that we probably don't want.
11:18So there's a couple phases that it wants to do here. So it's like a unified design system starts here. Uh, phase two to extract the component tokens, and then three is to fix the hard code values.
11:29And then four is to create a tailwind config for advanced customization, which is gonna be really nice. So now, it's asking me, would you like me to proceed with implementing this plan? I recommend we start with phase one.
11:39I should be able to look at slash context. And we're gonna get a visualization right now of how our context is being used. So as of now, with these tool calls, can kinda see it only used like less than a thousand tokens for all these different tool calls, which is amazing.
11:56So our MCP servers and tool calls only use a total of 2,800 tokens, which is only 1.4% of the context window.
12:04And you can see the value in just these two mcps alone on a refactoring task which requires touching a lot of files, reading them in, and, you know, doing this type of comparison work. I'm super impressed with this type of system and I found extremely good consistency combining the two.
12:21And let's just go ahead and check and see what's cursor is doing. I mean, cursor probably finished way earlier and you can see the token efficiency in cursor. He used 98,000 tokens.
12:29So one of the cool things about cursor's plan mode, which is what we didn't really compare here in Cloud Code, is the fact that it's now asking me a couple more questions. Like, should we handle semantic scope of brand colors and then the cream background? But the last thing we wanna go ahead and get set up here is getting this set up probably in codecs and also getting this set up in a fact, a droid factory.
12:48So for codecs, we're just gonna go ahead and go to what the instructions say for codecs. So in use ref here, they have like any MCP's clients.
12:56You can say find your NCP client for use ref, and we can say codecs here, and they'll say codecs CLI. So in codecs CLI, they give us the actual documentation reference here. And so for here, it says codecs is configured in a config dot toml.
13:10So if I go to terminal, and I just open up any tab, codecs m c p list.
13:19And we can see I currently have playwright installed. So I don't currently have the x o one installed. What I can do is I usually just copy this here and then go back to my terminal.
13:30You can even do this in cursor. It's probably just easier to just open up the terminal in cursor. So just in here, just type in cursor, cursor and then hit this.
13:40And this is actually gonna take us exactly to where this specific file is in our file system. So this one says to enable ref, update your TOML to include this. We're just gonna copy this.
13:49We're gonna go back in the cursor. We're gonna paste. And we still need to put our API key here from the use ref.
13:55So since we already had it configured in cursor, what I'm gonna go ahead and do is just scroll down to where it says tools MCP here on the side. And then I'm gonna go to ref and hit this little pencil. And so the pencil will already have ref listed with our API key.
14:07I'm gonna copy this API key and then in here in versus config dot toml where we're at, I'm just gonna go ahead and just delete this here and then paste it in here.
14:18I'm gonna go ahead and hit save and then now we should have it currently saved. So if I type in codecs MCP list, we should now be able to see that the MCP server is currently configured in here.
14:28So use ref MCP to look up the latest on tailwind v four and see if there are any gaps in my implementation.
14:41Okay. So we're just gonna hit enter. And so now it's gonna go ahead and send that task off and kinda do some research using the OpenAI Codex version.
14:50It's actually you can see here ref dot ref research documentation. This is actually now using the MCP that's built in. And so we're gonna do the same thing for Exa, and then we're gonna go back into here where it says Exa MCP.
15:01And then in here, we're gonna just type in like codecs. I'm gonna copy this URL since this already works inside of cursor. We're just gonna do the same thing and make it work inside of the config dot toml or codecs.
15:12And what I'm gonna go ahead and do is just paste this key right into here. So this is exactly the same key we had inside of cursor. And you can kinda see with that little question mark, that's a little parameter that we're gonna be passing in that shows the x API key.
15:22I'm gonna open up a new terminal window and say codecs MCP list. And now we should be able to see the exa MCP currently listed and status is enabled.
15:32If I paste this in, and then I should be able to get some more prompts or something back saying, hey, want to use this MCP and I just should be able to give us some approval. So for the factory droid, if you're using droid as an agent, what you wanna go ahead and do is kinda do something similar.
15:48I'm gonna show you real quick. So it's called cursor and then tld slash dot factory, and it's gonna be under slash m c p json.
15:59So when I do that, what what's gonna happen now is now we have this specific thing opened up inside of cursor. This allows us to actually edit the file and save it all in one place, which is really really nice. So all you have to do is get your configuration from cursor.
16:13And if you were getting it from cursor, all you do is go in here, hit tools MCP, go into this section, and all you have to do is literally copy this from this blue ref to here to this including this little blue line in that. And then you go back over here where it says MCP factory JSON right up there.
16:30And then you just paste it right in there and you hit save. Once you have these two MCP servers, I think what's gonna happen for your AI coding is that it's significantly gonna get better.
16:40But you actually have to call them specifically saying use ref MCP and then use exa MCP or use exa code for MCP. And so the other pro tip that I wanna give you before we kinda close out the video is the fact that you want to make sure that you use them in plan mode.
16:58So in any type of planning mode, you can usually hit shift tab once or twice. You'll see that all of these agents have like a spec mode, a planning mode of some sort. Make sure you use those specific keywords so that it gathers up all that specific context you want before you start writing code.
17:14This is at first getting started. If you want to learn a little bit more, I do have a couple more spots that are open for my one three three seven intensive. And so that's the elite if you don't understand what that means.
17:25And I can sit down with you for thirty minutes for five days in a row, and we're gonna try to get either MCP server set up, your documentation workflow set up, maybe something that's kinda stopping you from getting the results that you want. And I've been doing this type of stuff for several years with AI coding, but as far as engineering, I used to work at Apple as an engineer and I've been doing it for over a decade.
17:46And I have a lot of industry experience shipping software, having software that affects billions of people every single day. And if you're interested in that, feel free to check out the link in the description obviously.
17:57And I wanna leave you with this last thing. This AI coding thing is just starting to take off and we've thrown the to go in this really long baseball game of AI.
18:10And I'm really excited for this era of AI coding because this is now unlocking a lot more capability. And the models of what we're doing today are just getting off the ground.
18:21And these type of tooling like the XMCP and the useRefMCP, these are gonna be really essential to keep the language model training data, like the intelligence of the models here, and the training data that is currently missing, it's just gonna pull that right in and try to use this type of logic. These tools are really magical now because now they're available twenty four seven and they don't require bugging a senior senior engineer every single day.
18:44My name is Ray Fernando, and I'm really excited to be with you. And I'll see you in the next livestream. Peace out, y'all.
The Hook
The bait, then the rug-pull.
Documentation used to eat 100,000 tokens before a single line of code was written. Ray Fernando — ex-Apple engineer, decade-plus shipping production software — found the fix in two MCP servers that pull only the context the agent needs, precisely when it needs it. The proof is live on screen: a full Tailwind v4 design-token audit across a real codebase clocks in at 2,800 tokens — 1.4% of what Cursor burned on the same task.
Named ideas worth stealing.
00:15concept
Context Rot
The degradation of LLM output quality as the context window fills with broad, untargeted documentation — the agent gets dumb before it writes a line of code.
Steal forany video explaining why vanilla web search MCPs hurt more than they help
01:31model
Targeted Agentic Search
- ref.tools — indexed docs, no context bloat
- Exa AI — high-quality search built for coding tasks
- Explicit MCP calls in prompt
- Plan mode before code mode
Pull only the context needed for the specific sub-task, not everything that might be relevant. Combine ref.tools (documentation precision) with Exa (coding-task search quality) and always call them by name in the prompt.
Steal forCLAUDE.md rules section, any agentic coding workflow
How they asked for the click.
VERBAL ASK
17:50product
“I do have a couple more spots that are open for my one three three seven intensive.”
Soft close after delivering all value. Thirty-minute sessions over five days. Credentialed with Apple engineering background. Low pressure but well-timed after a high-value tutorial.
MENTIONED ON CAMERA
FROM THE DESCRIPTION
PRIMARY CTAWhere the creator wants you to go next.
OTHER LINKSAlso linked in the description.
Visual structure at a glance.
