Andrew Qu traces Vercel's data-science agent from a Snowflake-schema mega-prompt through chained specialists and a stateful agent, then shows why the decisive gains came from a sandboxed file system and roughly 100 skills distilled from recurring queries. The journey demonstrates how familiar tools and accumulated company knowledge can turn a fragile internal agent into a production system.
AI Engineer18 minTranscript found
Quick learning frame
Read this before watching.
Coding-agent workflow is the loop of inspect, route, plan, edit, verify, summarize, and decide what should be automated next.
New playlist item from AI Engineer; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Skill you build: The ability to evolve an internal agent around a concrete bottleneck, give it an explorable file-system workspace, and compound successful work into reusable organizational skills.
Watch for the shift from claim to mechanism. The learning value is the point where the transcript reveals a repeatable action, tool boundary, context move, review habit, or artifact.
Concept diagram
Where this video fits.
01Inspect context
02Route tool
03Plan work
04Edit safely
05Verify behavior
06Report next step
Deep lesson
Turn this video into working knowledge.
3,385 cleaned transcript words reviewed across 953 timed caption segments.
Thesis
How We Solved Agent Building — Andrew Qu, Vercel teaches a practical coding-agent workflow move: Andrew Qu traces Vercel's data-science agent from a Snowflake-schema mega-prompt through chained specialists and a stateful agent, then shows why the decisive gains came from a sandboxed file system and roughly 100 skills distilled from recurring queries. The journey demonstrates how familiar tools and accumulated company knowledge can turn a fragile internal agent into a production system.
The goal is not to remember the video. The goal is to extract the operating principle, tie it to timestamped evidence, test how far the claim transfers, and make something reusable.
0:58
Find the Bottleneck
“agents. And we've been embarking on a similar journey to make it easy for people to build agents and agentic applications easier. We built this thing called the AIDK. So instead of needing to switch out 300 400...”
Vercel's useful starting point was not a general-purpose agent but a concrete organizational constraint: its lean data team repeatedly stopped higher-value work to translate questions from marketing and sales into queries, analysis, and recommendations. Interview one team about the task they most dislike and map the repeated request, manual steps, and higher-value work that the interruption displaces.
5:04
Preserve Working State
“passes on a query to the planning agent that will then have an execution agent etc. And if you chain all of these together, you actually get something that looks like this where each agent has a very...”
Chaining narrowly scoped planning, SQL, execution, and reporting agents enabled an end-to-end loop, but each handoff exposed only a summary of prior work. Moving to one agent that managed its own state let it revisit earlier exploration when a query or join failed. Diagram a multi-step agent workflow and mark what context is lost at each handoff, then sketch how one stateful agent could recover from an execution error.
12:30
Files Build Memory
“the middle. And we thought, you know, building agents should be this simple. You should only have to create a skills folder, a tools folder, a channels folder, and you should be able to just declare these very...”
The largest improvement came from replacing prescriptive tools with a sandbox containing the semantic layer and familiar operations such as listing, reading, writing, and running files; this file-system agent doubled the eval score. Vercel then ran a recurring job that distilled repeated query patterns into roughly 100 skills, giving new runs company-specific context instead of making each one start from scratch. Run one repeated business question through a sandboxed file workspace, save the successful query pattern as a reusable skill, and test whether a fresh run can solve a similar question with less rediscovery.
01
Inspect context
Start with this video's job: Andrew Qu traces Vercel's data-science agent from a Snowflake-schema mega-prompt through chained specialists and a stateful agent, then shows why the decisive gains came from a sandboxed file system and roughly 100 skills distilled from recurring queries. The journey demonstrates how familiar tools and accumulated company knowledge can turn a fragile internal agent into a production system. Treat "Inspect context" as the outcome you are trying to make visible, not a topic label. Anchor it to 0:58, where the video says: “agents. And we've been embarking on a similar journey to make it easy for people to build agents and agentic applications easier. We built this thing called the AIDK. So instead of needing to switch out 300 400...”
02
Route tool
Use "Route tool" to locate the part of the coding-agent workflow mechanism the video is demonstrating. Ask what changes in your real setup if this claim is true. Anchor it to 5:04, where the video says: “passes on a query to the planning agent that will then have an execution agent etc. And if you chain all of these together, you actually get something that looks like this where each agent has a very...”
03
Plan work
Turn "Plan work" into the reusable artifact for this lesson: A coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal. This is where watching becomes something you can inspect and reuse.
04
Edit safely
Use "Edit safely" as the application surface. Decide whether the idea touches a browser flow, a local file, a model choice, a source document, a UI, or a review step.
05
Verify behavior
Use "Verify behavior" to prove the lesson. The evidence should connect back to the video title, transcript anchors, and a concrete output, not a generic best-practice claim.
06
Report next step
Use "Report next step" to carry the idea forward: save the prompt, checklist, diagram, or operating rule that would make the next agent run better.
Example
Source-backed artifact packet
Convert the video into a scoped artifact request that includes the transcript claim, mechanism, acceptance criteria, and proof. The output should be a coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal..
Example
Coding-agent workflow proof brief
Separate what the speaker claims, what the demo actually proves, and what still needs outside verification before you adopt the coding-agent workflow pattern.
Example
Teach-back module
Transform the lesson into a definition, a Inspect context -> Route tool -> Plan work -> Edit safely -> Verify behavior -> Report next step diagram, one misconception, one practice exercise, and a check-for-understanding question.
Do not learn it wrong
Treating the title as the lesson without checking what the transcript actually says.
choosing tools by hype
losing context across agents
letting parallel sessions become invisible
Letting the lesson drift into generic Codex vs Claude comparison.
Letting the lesson drift into feature lists without task routing.
Letting the lesson drift into claims that ignore limits or recovery.
Do not count this as learned until these are true.
01
State the transcript-backed claim in your own words: Andrew Qu traces Vercel's data-science agent from a Snowflake-schema mega-prompt through chained specialists and a stateful agent, then shows why the decisive gains came from a sandboxed file system and roughly 100 skills distilled from recurring queries. The journey demonstrates how familiar tools and accumulated company knowledge can turn a fragile internal agent into a production system.
02
Explain the practical stakes without hype: New playlist item from AI Engineer; queued for transcript-backed review, topic mapping, and a practical learning artifact.
03
Map the idea onto the Inspect context -> Route tool -> Plan work -> Edit safely -> Verify behavior -> Report next step sequence and name the weakest link.
04
Produce the artifact and include the evidence that proves it: A coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal.
Put it into practice
Give this grounded prompt to Codex or Claude after watching.
You are helping me turn one specific YouTube video into real, durable learning.
Source video:
- Title: How We Solved Agent Building — Andrew Qu, Vercel
- URL: https://www.youtube.com/watch?v=9dYcwOkpCE8
- Topic: Codex + Claude Workflows
- My current learning frame: Choose one recurring internal request, prototype it in a sandbox with a small set of file tools, then distill the successful procedure into a reusable skill and test that skill on a similar fresh request.
- Why this matters: New playlist item from AI Engineer; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Transcript anchors from this exact video:
- 0:58 / Evidence 1: "agents. And we've been embarking on a similar journey to make it easy for people to build agents and agentic applications easier. We built this thing called the AIDK. So instead of needing to switch out 300 400..."
- 5:04 / Evidence 2: "passes on a query to the planning agent that will then have an execution agent etc. And if you chain all of these together, you actually get something that looks like this where each agent has a very..."
- 8:43 / Evidence 3: "really just use a file system. You know, we we saw the learnings from claw code and how powerful it was given that it just executes locally. And we tried to rebuild it in a way that was..."
- 11:00 / Evidence 4: "layer and the system prompt. But with a skill, it already starts off with a lot of contextual knowledge that has otherwise already been done. And this is roughly how it looks. It's very similar to the previous..."
- 12:30 / Evidence 5: "the middle. And we thought, you know, building agents should be this simple. You should only have to create a skills folder, a tools folder, a channels folder, and you should be able to just declare these very..."
- 14:22 / Evidence 6: "a mini claw to go and test people's services. It goes to websites, installs them, it tries to use them. And they've seen incredible success on building their own agent from the ground up using Eve compared to..."
- 15:57 / Evidence 7: "a squeeze, you should really try to build your own agent and add in as much company specific knowledge as you can. Today, you know, we've had 20 roughly decently PMF agents adversel that range from anything from..."
Video-aware target:
- Prompt lane: Coding-agent workflow
- Mechanism to extract: Find the workflow rule that explains when and how to use Codex, Claude Code, browser control, dashboards, or manual review.
- Artifact to produce: A coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal.
- Artifact must include: task class; agent/tool choice; context packet; verification step; handoff/recovery rule
Your task:
1. Use the transcript anchors above as the primary source packet. If you add outside context, label it clearly as outside context and keep it secondary.
2. Create a source-check table with columns: timestamp, claim, transcript support, what the demo proves, confidence, and what still needs verification.
3. Extract the actual teachable mechanism from the video: Find the workflow rule that explains when and how to use Codex, Claude Code, browser control, dashboards, or manual review. Do not invent claims that are not supported by the title, lesson frame, or transcript anchors.
4. Build a reusable learning artifact: A coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal.
5. Include:
- a plain-English definition of the core idea
- a diagram or structured model using this sequence: Inspect context -> Route tool -> Plan work -> Edit safely -> Verify behavior -> Report next step
- answers to these source questions: What workflow pain is being solved? | What exact coordination mechanism is shown? | What evidence proves it changes the work?
- 3 concrete examples that apply the video idea to real agentic work, such as rate-limit routing; browser verification after a UI edit; long-running /goal session review
- 2 failure modes the video helps prevent, chosen from the transcript evidence and these likely risks: choosing tools by hype; losing context across agents; letting parallel sessions become invisible
- a checklist for the next real workflow, focused on: routing decision, context portability, verification, handoff summary
- one practical exercise with a clear done signal: Route three recent tasks across Codex, Claude, browser checks, and manual review with a reason for each.
6. Add a "learning transfer" section: what changes in my workflow tomorrow if I actually learned this?
7. Add a "source check" section that cites which transcript anchor supports each major takeaway.
Quality bar:
- Make this specific to "How We Solved Agent Building — Andrew Qu, Vercel", not a generic Codex + Claude Workflows essay.
- Each workflow rule must point to a timestamped claim or demo moment, then state what remains unproven.
- Prefer operational examples, failure modes, and reusable artifacts over broad definitions.
- Call out uncertainty instead of smoothing over weak evidence.
- Avoid these generic drifts: generic Codex vs Claude comparison; feature lists without task routing; claims that ignore limits or recovery.
- If evidence is weak or missing, stop and say what transcript segment or timestamp needs review instead of guessing.
- Finish with a concise artifact I could paste into my learning app.
Misconceptions
What to stop believing.
One agent should do every task.
Different tools have different strengths. Routing is part of the workflow.
More context is always better.
Relevant context helps; stale context causes drift and cost.
Practice studio
Learning only counts when you make something.
01
Transcript evidence map
Separate what the video actually says from what you already believe about the topic.
3 source-backed takeaways with timestamps, confidence, and a transfer note.02
One useful artifact
Apply the video to a real workflow and produce a coding-agent routing and execution matrix with context needed, tool choice, verification, and done signal..
A reusable artifact with a done signal and one verification step.03
Coding-agent workflow teach-back card
Explain the coding-agent workflow mechanism to someone who has not watched the video yet.
A 90-second explanation, one diagram, one example, and one misconception to avoid.
Recall check
Answer first, then reveal — without rewatching.
Why did Vercel choose the data team's query workflow for its early internal agent experiment?
What advantage did one state-managing agent have over the earlier chain of specialized agents?
Why did the file-system design and distilled skills improve D0?
Source shelf
Use the video as a doorway, then verify with primary sources.
