aranculleton/last-job-search / .claude/skills/job-application-assistant
job-application-assistant
VerifiedAssists with job applications: evaluating job postings, tailoring resumes, writing cover letters, verifying the compiled PDFs, and preparing for interviews. Triggers on keywords like: job posting, job application, CV, cover letter, resume, interview prep, job fit, career, application, apply, tailor, ATS, shortlist, build my resume, build one with me, no resume, show me around, what can you do, how do I use this
| File | Kind | Size |
|---|---|---|
| 03-writing-style.md | markdown | 24.3 KB |
| 04-job-evaluation.md | markdown | 34.5 KB |
| 05-cv-templates.md | markdown | 50.9 KB |
| 06-cover-letter-templates.md | markdown | 11.3 KB |
| 07-interview-prep.md | markdown | 12.5 KB |
| 08-application-forms.md | markdown | 7.9 KB |
| 09-web-research.md | markdown | 10.6 KB |
| 11-cost-model.md | markdown | 10.9 KB |
| SKILL.md | skill md | 8.0 KB |
| setup-sections.md | markdown | 26.7 KB |
| verification-checklist.md | markdown | 8.3 KB |
07-interview-prep.md
View on GitHub---
framework_version: 1.1.6
---
# Interview Preparation Guide
The method is the app's; the examples are not anybody's. The STAR examples below belong to
Morgan Ellis, the fictional person in `profile.example/`, and show the shape only. A real
person's are in `profile/07-interview-prep.md`, which /setup writes and `/interview` reads.
Never use these for a real person, and never edit this file for one.
## STAR Format
Structure answers as: **Situation** (context), **Task** (your responsibility), **Action** (what you did), **Result** (outcome).
Keep answers to 1-2 minutes. Be specific. End with what you learned or would do differently.
## Ready-Made STAR Examples
<!-- Worked examples for Morgan Ellis, the fictional example person in profile.example/.
A real person's own, drawn from their resume, are in profile/07-interview-prep.md. In a real set every
number appears on the resume, except a draft the person asked for, which is marked for them to
confirm (/interview rule 2); do not inflate any of them in the room. Where an interviewer pushes
for detail the resume does not carry, say what you actually remember and flag the rest as
approximate. -->
### 1. The forecast that freed safety stock (applied modelling)
**S:** Weekly demand forecasts for 2,400 SKUs across six distribution centres ran at a 22% error, and planners were covering the gap with extra safety stock.
**T:** Own a weekly forecast the planners would trust, from the raw warehouse data to the number they order against.
**A:** Rebuilt the forecasting models in Python and SQL with promotions and seasonality as explicit features, and tracked error by distribution centre so the planners could see where the number held.
**R:** Forecast error down from 22% to 14%, and an estimated £1.1M of safety stock released.
**Use for:** "Walk me through an end-to-end project", "How do you validate a forecast", "Tell me about a model you built", "How do you make a model actually get used"
### 2. Ending a daily manual reconciliation (initiative and automation)
**S:** Every morning someone spent two hours reconciling the warehouse-management data feed against the reports that depended on it, and nobody owned fixing it.
**T:** Remove the manual step rather than keep absorbing it.
**A:** Rebuilt the feed as scheduled dbt models with data-quality tests, so a bad load failed loudly instead of surfacing in a report.
**R:** The daily two-hour reconciliation ended.
**Use for:** "Tell me about a time you took initiative", "How do you improve a process", "Give an example of building something reusable", "How do you test data pipelines"
### 3. The route-consolidation model (commercial impact)
**S:** Regional deliveries were planned route by route, and vans were leaving partly empty.
**T:** Build a model the transport planning team would actually use, not one that only looked good in a deck.
**A:** Designed a route-consolidation model for regional deliveries, walked the planners through how it chose each consolidation, and adjusted its constraints to match how the depots really loaded vans.
**R:** 9% of weekly van miles removed, and the model adopted by the transport planning team across the North West.
**Use for:** "Tell me about commercial impact", "Describe presenting technical work to a non-technical audience", "How do you get a team to adopt your work", "Tell me about a model that changed an operational decision"
### 4. The replenishment-rules pilot (evidence before rollout)
**S:** Oakmere Foods had revised replenishment rules that planners were confident in, with no evidence yet that they improved availability.
**T:** Show whether the rules worked before they went chain-wide.
**A:** Designed a pilot in 12 stores against matched control stores, fixed the availability measure before the pilot started, and evaluated the result with the uncertainty stated.
**R:** A significant availability gain, and the rules went chain-wide.
**Use for:** "Tell me about a decision you backed with data", "How do you test something you cannot fully randomize", "Describe influencing without authority", "How do you handle a result people did not expect"
### 5. Replacing eleven spreadsheets with one dataset (stakeholders and trust)
**S:** Thirty-plus operations managers were working from eleven different spreadsheets for service levels, and the numbers disagreed.
**T:** Give them one definition of service level and one place to read it.
**A:** Built a governed dataset and the Power BI reporting on top of it, agreed the definitions with the managers one by one, and retired the spreadsheets as each team moved over.
**R:** One service-level view used by 30+ operations managers, in place of eleven spreadsheets.
**Use for:** "Tell me about managing stakeholders", "How do you handle conflicting numbers", "Describe getting people off a tool they like", "What does a single source of truth mean to you"
### 6. Chilled-line write-offs (statistical modelling for a commercial result)
**S:** Chilled lines were losing stock to expiry, and ordering rules took no account of shelf life.
**T:** Understand how shelf life and order patterns interacted, and turn it into a change the buyers could make.
**A:** Modelled shelf-life and order patterns for the chilled lines and gave the buyers specific order-quantity changes rather than a general finding.
**R:** Write-offs fell 18% over two quarters.
**Use for:** "Tell me about a time analysis changed a business result", "How do you turn a finding into an action", "Describe working with commercial teams", "Give an example of a measurable outcome"
<!-- Add more STAR examples as needed. Aim for 4-6 covering different competencies. -->
## Common Tough Questions
An answer can end on a `Suits:` or a `Not for:` line naming the kinds of employer it fits or argues against (consulting, product, bank, public sector, startup). `tools/interview_pack.py` leaves out a stock answer that argues against the employer it is about to be said to, "depth over rotating clients" said to a consultancy, says why, and lists it under what not to say; it reads untagged answers for the words that would.
### "Why are you leaving Brindle & Hart Logistics?" / "Why are you looking?"
> I am not unhappy - the role has given me ownership of the forecasts from the warehouse data feed through to what the planners order. The move is about scope. I have taken demand forecasting about as far as one distribution network allows, and I want a larger or more varied planning problem, with a team that plans over a longer horizon than the week. `[CONFIRM: check this matches your actual reason before using it.]`
### "Do you have the right to work here, and will you need sponsorship?"
> Answer both plainly, exactly as the work-authorization facts in `profile/01-candidate-profile.md` record them. Where the profile says no sponsorship is required, the second is a plain no, with no caveat. `[PLACEHOLDER: only if the candidate's status has an expiry date, add the answer for when they are asked about it directly. Follow the disclosure rule their profile records: many people say only that they are entitled to work and need no sponsorship, and name the permit or its dates only when asked. Asked, say it plainly and say what the plan is, because evasiveness reads worse than the status itself. Never volunteer it. Delete this note otherwise.]` `[CONFIRM: check the facts against the profile before any interview.]`
### "Walk me through your career - why these moves?"
> Each role took on more of the chain than the last: a graduate year forecasting footfall for retail clients at Saltmarsh Analytics, then in-house at Oakmere Foods owning the measurement on stock and suppliers, then Brindle & Hart Logistics owning the forecasts and the data feed behind them. What I am looking for now is a larger planning problem to own. Say it in one breath and move on.
### "What were you doing between [date] and [date]?" (use only if the profile has a gap)
> `[PLACEHOLDER: two sentences stating what the period actually contained, drawn from 01-candidate-profile.md. Delete this question if there is no gap.]`
>
> **Delivery notes.** Say it in two sentences and stop; a gap sounds like a problem only if you narrate it like one. Land the strongest true fact plainly and do not oversell it. Mention a minor job only if asked what it was, and never put it on the resume or claim it as professional experience. If pressed, be matter-of-fact.
### "You don't have [specific skill/experience]."
> Name the gap plainly, then bridge to the nearest real thing and say what you are doing about it. Worked examples: no SAP configuration, but I have reported from SAP extracts and the warehouse-management system behind them. No Airflow, but I have run scheduled dbt models with data-quality tests in production. No people management, but I have owned reporting used by 30+ operations managers and moved them off eleven spreadsheets. No experience in your specific sector, but I changed sector between each of my roles, and each time I learned the commercial mechanics fast enough to be useful in the first month. Unasked, never claim the skill; asked for an example of using it, draft one, marked for the person to confirm (/interview rule 2).
### "Where do you see yourself in 5 years?"
> Leading the planning analytics for a network end to end - the forecasts, the data feeds behind them and the relationship with the planners who act on the output - with a small team around it. I would rather go deep on planning than broad across many analytics topics.
Not for: consulting
### "What's your biggest weakness?"
> I tend to polish a forecast before showing it to the people who will use it, which delays the feedback that would have improved it. At Brindle & Hart I now share a rough weekly version early and agree the error measure with the planners before I tune anything. It costs a little pride at the start and saves rework later, but I still have to push myself to show work early. `[CONFIRM: check this matches your actual experience before using it.]`
### "Why this company specifically?"
> Customize per company. Must reference: specific projects, company values, market position, or team structure. Never give a generic answer.
## Questions You Should Ask Interviewers
### About the Role
- "What does a typical week look like in this role?"
- "What would success look like in the first 6 months?"
- "What's the biggest challenge the team is facing right now?"
### About the Team
- "How big is the team, and how do you divide work?"
- "How does a piece of work go here, from idea to production?"
- "How do you onboard new team members?"
### About Tech & Growth
- "What's your current tech stack for [relevant area]?"
- "Is there room to grow into bigger, more strategic decisions?"
- "How does the team stay current with new tools and methods?"
### About Culture (use these to prevent disappointment)
- "How would you describe the team culture?"
- "What does professional development look like here?"
- "Is there flexibility to work from home some of the week?"
- "What's the balance between new work and keeping things running?"
- "How would you describe the leadership style in this team?"
- "What do people who thrive here have in common?"
## Phone/Video Interview Tips
- Have your STAR stories written out (profile/07-interview-prep.md; the ones in this file are the example person's)
- Keep a glass of water nearby
- Smile when speaking (it changes your tone)
- Ask for clarification if a question is vague
- It's OK to take 5 seconds to think before answering
- End with: "Is there anything about my background you'd like me to go into more?"
## After the Application (Best Practice)
### Follow-Up Etiquette
- **Don't call to "stand out"** or to learn more about the role post-submission - this risks a negative impression
- If the employer specified a timeline, respect it and wait
- If no timeline was given and significant time has passed (2+ weeks), a brief call to ask about status is acceptable
- If you have genuinely new, relevant information to share, a short follow-up is fine
### Thank-You Notes
- When you receive any update (interview invitation, rejection, or status update), send a brief thank-you message
- Express appreciation for their time and the process
- Keep it short (2-3 sentences)
## Roleplay Guidelines
When the user asks for interview practice:
1. Ask which role/company to simulate
2. Start with easy warm-up questions ("Tell me about yourself")
3. Progress to role-specific technical questions
4. Include 1-2 behavioral questions using the competencies from the job posting
5. End with a tough question or curveball
6. After each answer, give brief feedback: what worked, what to sharpen
7. Suggest which STAR example would work best for each question
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |