The Ask Behind the Ask
You are a forward-deployed engineer. Yesterday your company signed a six-week paid pilot with Meridian Freight Systems, a less-than-truckload freight brokerage — $410M in revenue, 340 brokers across six service centers, 1,400 employees…
Create a free account to upload your work. Your progress saves as a draft until you submit.
What You'll Be Doing
You are a forward-deployed engineer. Yesterday your company signed a six-week paid pilot with Meridian Freight Systems, a less-than-truckload freight brokerage — $410M in revenue, 340 brokers across six service centers, 1,400 employees total.
Below is an excerpt from this morning’s kickoff call, and the fact sheet their IT director sent afterward.
Dana Whitfield, SVP Operations: What I want is simple. A shipper emails us asking for a rate. Right now it takes us forty-seven minutes on average to get back to them. I want the AI to just answer it. Instantly.
You: When you say answer it —
Dana: Send the quote. Customer emails, the AI reads it, the AI sends the price back. My brokers stop typing and start selling.
Priya Raghunathan, Director of IT: Rates don’t live in one place, Dana. Contract rates are in Kestrel. Spot rates come off three carrier portals that have no API. And about a third of our lane pricing is a spreadsheet Ron in Dallas maintains.
Dana: Ron’s spreadsheet is the best pricing we have.
Priya: It’s a spreadsheet on a shared drive. And nobody is writing to Kestrel from outside Kestrel. Not in a pilot. Not in production either, probably.
Dana: Look. Delcourt is quoting our shippers back in nine minutes. Nine. If we’re at forty-seven we are not in the conversation. I need something in six weeks I can put in front of the board.
Fact sheet (from Priya, same day):
| Item | Detail |
| Inbound to ops@ | ~11,000 emails/day, of which ~2,300 are quote requests |
| Kestrel (their TMS) | In-house, 14 years old, SQL Server. REST API added 2021, covers ~60% of entities |
| Kestrel write access | None for external systems. Read replica only, refreshed every 15 minutes |
| Contract rates | 4,100 lanes, in Kestrel |
| Spot rates | Three carrier portals. Screen-scrape only, no API |
| “Ron’s spreadsheet” | One XLSX, 7 tabs, ~2,900 rows, updated by hand roughly weekly |
| Win rate, quote returned in under 10 min | 31% |
| Win rate, quote returned in over 30 min | 12% |
Write the scoping note you would send your account lead tonight.
Constraints to consider
- No write access to Kestrel during the pilot. You get the read replica, 15-minute lag. Anything you build sits beside their TMS, not inside it.
- Everything runs inside Meridian’s AWS account. No customer data leaves their VPC, and you cannot introduce a service that is not already on their approved vendor list.
- You get four hours total of Priya’s engineers over the whole six weeks, and they must be scheduled in advance. Anything you need from IT has to be named up front, in this note.
- Nothing goes to a shipper without a named human approving it. Meridian’s contracts with their shippers make an auto-sent price a commitment.
- Your note is one page and names exactly one first deliverable for week 1. Not a phased roadmap. One thing.
AI Usage Guidance
We expect you to use AI tools. We evaluate how you use them — not whether you use them. Evidence of iteration, redirection, and critical evaluation scores higher than a polished output with no process documentation.
The single highest-signal indicator: your video answer to the mandatory AI question. If you cannot name a specific moment where you redirected AI output, evaluators will assume you did not.
Mandatory AI question for your video: Walk me through one moment where you disagreed with, pushed back on, or redirected what the AI gave you — and what you did instead. Name the specific moment. Explain what the AI produced that didn’t meet the bar, what you did differently, and why.
Speak naturally. Communication is assessed on clarity of technical ideas and logical structure — not verbal polish, accent, or filler words.
Submission: Upload each deliverable as a separate file directly on the Provn platform.
What You'll Accomplish
Separate what a customer asked for from what would actually move their number
Bound a six-week engagement down to a single week-1 deliverable and justify the cut
Identify a project dependency that no one in the room owns, and treat it as risk
Convert a fixed, scarce customer-engineering budget into a specific, scheduled ask
Explain a scoping decision to a customer-facing colleague in under four minutes
How Your Work Will Be Scored
What to Submit
File 1 — Scoping Note
Format: .pdf, .doc, .docx, .rtf, .txt, .md
What Dana actually needs versus what she asked for. The one thing you build in week 1. What you need from Priya, stated as something she could put on a calendar.
Sign in to upload files
File 2 — README Document
Format: .pdf, .doc, .docx, .rtf, .txt, .md
Three required sections:
- Section A — Written analysis: the risk you are most worried about in this engagement, and the cheapest way to find out in week 1 whether it is real. 150–250 words.
- Section B — The unblocking question: the single question you would ask before writing any code, and — for each plausible answer — what it changes about your plan. Give at least two branches.
- Section C — AI Usage Log (Mandatory): This is not a trick. We want to see how you work with AI — not whether you used it. In a short section of your README, document your AI collaboration process. For each significant interaction with an AI tool, briefly note: what you asked the AI to help with / what it gave you / what you kept, changed, or rejected — and why. Three interactions documented is sufficient. The log does not need to be exhaustive.
Sign in to upload files
File 3 — Video Walkthrough
Format: .mp4, .mov, .webm
Record as MP4 or MOV and upload directly on the Provn platform as a separate file. No ZIP files, no repository links, no third-party video platforms.
Cover: (1) what you think Dana actually needs and how that differs from what she said, ~60 sec; (2) your week-1 deliverable and the dependency you don’t control, ~60 sec; (3) the mandatory AI question, ~60 sec; (4) the one question you’d ask on the next call, ~30 sec. Speak naturally — we’re assessing your thinking, not verbal polish.
Sign in to upload files
Create a free account to upload your work. Your progress saves as a draft until you submit.
On this page