Clear answers before you trust a CV tool with your applications.

Asking ChatGPT to rewrite your CV can be useful, but it is still usually one conversation with one model trying to do everything at once. This service is built as a structured 7-step system designed specifically for job applications, with separate stages for analysis, rewriting, critique, authenticity review, scoring, packaging, and final delivery.

Behind the scenes, that workflow is coordinated through our own MCP-based orchestration layer and uses three different AI models for three different jobs. One model handles the first rewrite against your CV and the target role. A second model reviews that output like a skeptical hiring manager and looks for weaknesses. A third model checks whether the result sounds natural, grounded, and genuinely human rather than polished in a generic AI way.

That separation matters. A model that writes something is not always the best model to judge whether it overstates your experience, copies the language of the job ad too closely, or slips into flat, formulaic phrasing. By splitting those responsibilities, the system can challenge its own output instead of simply accepting the first decent-looking draft.

There is also a cost and effort difference. In theory, you could try to recreate something similar manually with ChatGPT, extra prompts, repeated rewrites, and perhaps multiple tools. In practice, that usually means more trial and error, more time spent steering the output, and often more money if you are paying for premium AI access or additional model usage. Most people do not want to spend an evening acting as the quality-control layer for their own CV.

The goal here is not just to generate text. It is to give you something closer to a checked, role-specific application workflow that has been built to produce a stronger final result with less manual prompting from you.

That concern is the reason the authenticity check exists as a dedicated pipeline stage, not a prompt instruction.

A third model - operating separately from the one that generated your content - scores the output across 7 specific dimensions. Language authenticity checks for formulaic sentence structure and repetitive openings. Cliché elimination checks buzzword density. Human reading test checks whether the rhythm and phrasing reads as produced by a person. Structural naturalness checks for mechanical paragraph uniformity. All 7 must score 0.8 or above.

The model doing this check did not write the content. It cannot be compromised by the original generation context. When dimensions fail, the specific violations and suggested fixes are passed to a revision step as typed objects. The revision addresses each failure explicitly before the output reaches you.

There is also a banned-phrase list enforced at generation: "results-driven," "team player," "proven track record," "passionate about," "leverage my expertise," and several dozen others are excluded at the source, not caught after the fact.

Rewritten CV - every work experience entry rewritten against the target job description. Every claim anchored to what you originally documented. A separate verification step traces each bullet and theme back to its source section in your CV. If a claim cannot be supported, it is not included.

Full change log - every edit documented with the specific reason it was made. What changed, what was added, and why. You can review each modification and revert anything you disagree with.

ATS compatibility score - your rewritten CV scored across five dimensions: keyword coverage, competency alignment, experience relevance, quantification, and structural compliance. Missing keywords, gaps, and formatting issues identified with specific fixes before you submit.

Three cover letter styles - a direct professional pitch; a problem-framing version that opens with the company' stated challenge; a career narrative tracing your progression toward the role. Each is structurally distinct, each passes the 7-dimension authenticity checklist.

Interview prep Q&A - anticipated questions based on the role and your documented background, with suggested response framings.

All outputs are returned in structured form and stored against your account for reference.

No. Before any model call, a dedicated sanitization step removes names, email addresses, phone numbers, URLs, and LinkedIn references from your CV text. The models receive a document with no identity attached. This is a technical control: there is nothing to transmit because the data is removed before transmission occurs.

Your real identifiers are reinserted locally in your browser after the AI response returns. A PII map is maintained client-side for this purpose. The models and the server never hold the mapping between redacted placeholders and your actual information simultaneously.

Full data handling policy →

The uploaded file and the text extracted from it are held in memory during processing only. When the request completes, they are discarded. No file storage. No database entry. No reconstruction possible.

What is stored is the generated output - the rewritten CV and cover letters - associated with your account. That stored record does not include your original CV, the job description, your personal identifiers, or the intermediate AI inputs. It is the finished document only.

What this service cannot do with your data, by design →

A model optimized for generation is not well-suited to detecting its own failures. The separation is structural, not cosmetic.

The first model receives your anonymized CV and the job description. It produces the initial rewrite - mapping experience to requirements, identifying transferable skills, rewriting bullets with specific claims.

The second model receives only the output of the first - not the original CV, not the job description, not the generation instructions. It operates as a hiring manager reviewing a document cold. It cannot be influenced by the reasoning that produced the content. It checks against 10 named failure categories and returns machine-readable findings with severity levels.

The third model receives only the output that passed critique. It checks for AI-generated patterns across 7 scored dimensions. It knows nothing about the generation or the critique - only the document in front of it.

If critique or authenticity fails, a revision step runs against the specific typed failure objects. The maximum number of revision loops is bounded. If the pipeline does not fully pass within that limit, partial outputs and structured failure reasons are returned - not silent failure and not an unchecked result.

Full technical architecture →

A dedicated step traces every claim in the output back to its source in your original CV. For each bullet and theme, it identifies: the exact source CV section, the original text, how it was transformed, and whether it is supported. Unsupported claims are flagged separately.

Evidence rules are enforced throughout. Metrics that do not appear in or cannot be directly derived from your CV are not used. "~50 users" cannot become "thousands of users." If no metric exists, scope is described specifically without fabricating a number. Seniority framing cannot exceed what your documented roles and dates support. If the fit between your background and the job is partial, the output reflects that narrowly rather than overstating it.

No. Everything is pay per optimization.

That is a deliberate choice. A lot of similar services pull people in with a free trial, then convert them into recurring charges they did not mean to keep. People forget to cancel, get billed again and again, and sometimes end up paying for months or even years longer than they planned.

We do not want that relationship with our users. The goal here is to help you find a job, not lock you into a subscription or make money from people who forgot a billing date.

• Starter Pack: €4.99 - 3 credits for 3 CV revisions + 9 cover letters
• Popular Pack: €13.99 (€1.40 each) - 10 credits for 10 CV revisions + 30 cover letters
• Pro Pack: €29.99 (€1.20 each) - 25 credits for 25 CV revisions + 75 cover letters

No recurring billing. No free trial that quietly rolls into a paid plan. One credit covers one role-specific optimization run against the job description you paste, and any unused credits stay on your account for later.

That means you can come back when you need to tailor your CV for another role, or use those credits later when helping a friend with their application materials. You pay when you need the service, and you keep what you bought.

Full pricing breakdown →

PDF only, including scanned and image-based PDFs. For image-based files, OCR runs server-side to extract text before the pipeline begins. Only the extracted text - after PII stripping - proceeds to the models. Raw image data is not transmitted.

Processing is handled as a background job. On average it takes around 4 minutes from submission to results. In some cases - typically when your CV has a lot of detail or the job description is dense - it can run up to 10 minutes. You do not need to stay on the page for this. See the next question.

Yes. When your optimization is complete, we send a notification email to the email address tied to your account so you know your result is ready.

That is intentional. Job searching is stressful enough without having to sit and stare at a progress screen. You can submit your CV, get on with your day, and come back when the work is done.

No. API calls to the AI providers are made under commercial terms that prohibit use of API request data for model training by default. The content you submit for processing is not used to improve any AI model.

Additionally, the content sent to the models has already had your personal identifiers removed. Even under the provider' default data handling, the models do not receive information that could identify you.

The pipeline handles this internally. If critique or authenticity findings exceed the pass threshold, a revision step runs automatically against the specific typed failure objects - not a vague re-prompt. The revision addresses each named issue before rechecking.

The number of revision loops is bounded. If the output does not fully pass within that limit, you receive whatever was produced along with the structured failure reasons - which failures occurred, in which dimensions, and what the specific violations were. You are never returned an unchecked result without being told it failed to pass.

In practice, the majority of outputs pass within the first or second revision loop.

No. Once you submit, the job is queued and runs in the background. You can close the tab, navigate somewhere else, or shut the browser entirely - it makes no difference. When you come back and log in, your results will be there waiting for you.

Nothing is cancelled if you leave. The pipeline runs through all its stages regardless of whether you are watching, and when it finishes we email the registered address on your account to let you know the result is ready.

In other words, once your submission is in the system, you do not need to babysit it. You will not lose your credit by closing the tab mid-process.

First, your CV is prepared for processing and personal identifiers are stripped before any AI stage begins. Then the system analyzes your background against the job description, generates tailored application materials, reviews them, checks whether they sound natural, and saves the final result to your account.

Some of those tasks can run asynchronously in the background to keep the experience smoother. Others have to run one by one because each stage depends on the result of the previous stage. That is why the process behaves more like a carefully managed workflow than a chatbot giving you a one-shot answer.

When everything is complete, your result is stored in your account history and we send you an email so you know it is ready.

Because what happens behind the scenes is substantially more than a single prompt sent to a model.

When you submit, a multi-stage pipeline kicks off. Some parts can run in parallel to save time, but the quality-control steps have to run in sequence because each one depends on the output of the stage before it. That includes generation, critique, authenticity review, ATS scoring, and the final packaging of the result.

If something is too generic, too weak, too artificial, or not specific enough, we do not just shrug and send it anyway. The content is revised and checked again. Those loops are a big part of why the output is stronger than what you get from instant tools.

We also use our own internal MCP server and orchestration layer to coordinate parts of that workflow cleanly. That sophistication is not there to sound impressive. It is there because a better system design leads to a better customer experience: more reliable processing, better context handling, and stronger final outputs.

So yes, it takes longer. But the extra time is the quality control doing its job, not the system sitting idle.

Because we want the system behind the product to be more sophisticated than a basic one-prompt workflow.

Our MCP server is part of the internal infrastructure that helps coordinate the different stages of the optimization process. In practical terms, that means we can structure the workflow more carefully, pass the right context into the right step, manage retries when something is not good enough, and keep the whole pipeline more consistent from start to finish.

The benefit to you is not just technical elegance. It is a better customer experience. More reliable processing, more thoughtful outputs, cleaner handling of multi-step work, and less chance that you are being handed the first draft and told it is finished.

We use a mix of leading models from Anthropic, OpenAI, and Google - specifically Claude, GPT, and Gemini - and we keep the stack current as better production-ready versions become available.

Different parts of the workflow benefit from different strengths. One model may be best for the first rewrite, another for stricter critique, and another for authenticity and pattern detection. We choose models based on output quality, not because one provider needs to handle everything.

We also do not optimize this product around squeezing out the highest possible margins by using the cheapest possible processing. The goal is to give you the strongest final output we can justify, even when that means running multiple high-quality models and extra validation steps behind the scenes.

In plain English: we would rather spend more on getting the work right than cut corners and give you something faster but weaker.

Because the output is not accepted just because one model produced it.

We use a multi-step process where generation, review, and authenticity checking are separated. That matters because a system that writes content is not always the best system to judge whether that content is strong, believable, and specific enough.

If the draft sounds generic, overstates your experience, or reads like obvious AI, it is not treated as final. It gets revised before it reaches you. The goal is not just to produce something quickly. The goal is to produce something you can actually send with confidence.

No. We do not fabricate achievements, tools, responsibilities, or experience that are not real.

What the system can do is notice that certain tasks in your CV often imply related experience that you may simply not have written down clearly. In those cases, it may surface that as a suggestion or a question back to you - for example, asking whether you also used a particular tool, handled a related responsibility, or worked in a way that is consistent with what is already described.

That is an important distinction. We may help you uncover relevant experience you already have but did not mention well, yet we do not turn assumptions into facts. Anything uncertain should be treated as something for you to confirm, refine, or reject.

The aim is to present your real background more clearly and more strategically, not to turn you into a different candidate on paper.

No, and that is deliberate.

If we focused on templates, we would end up helping create large numbers of CVs that look very similar. That is not our specialty, and it is not where we think we add the most value. There are already plenty of free and paid template libraries online for people who want a different visual layout.

What matters more today is that your CV is readable, clearly structured, and easy for automated systems to process. In many hiring pipelines, your CV is scored, summarized, or filtered before a human gives it real attention. Hiring teams often see keyword matches, fit summaries, ATS signals, or condensed notes before they ever read the full document closely.

That is why we focus on the substance of the CV first: the wording, the evidence, the alignment to the role, and the clarity of what you have actually done. A beautiful template cannot rescue weak positioning, but strong positioning can work well in a clean, readable format.

The pipeline still runs, but the quality of tailoring is directly proportional to the detail available. A one-line job posting gives the models very little to anchor against. A full description with responsibilities, required skills, and context about the team or product gives the system far more to work with.

If you have access to more information about the role - from a company page, a recruiter conversation, or a LinkedIn posting with fuller detail - pasting that alongside or instead of a sparse job ad will produce meaningfully better output. The models use whatever you give them; there is no background research done on the company or role.

Because it is usually the wrong tradeoff for a service like this.

First, there is the legal and terms-of-use side. Many job boards and application platforms place clear restrictions on automated scraping, and we do not want to build a product that depends on operating in a grey area or pushing against rules other companies have set. If a user pastes the job description themselves, the source is explicit and the process is much cleaner.

Second, scraping sounds simple from the outside, but in practice every site is different. Layouts change, anti-bot protections change, application flows differ, and what works on one site may fail on another a week later. That creates a lot of maintenance work for something that does not actually improve the quality of the CV rewrite itself.

Third, doing this through browser agents or more advanced automation would increase processing cost. That extra work means more infrastructure, more model usage, and more points of failure, which would push the price up for everyone. We would rather keep the service focused on the part that matters most: taking the job description you provide and turning it into a strong, well-tailored application.

So the short version is this: asking you to paste the job description is more reliable, more respectful of platform rules, and better for keeping the service affordable.

The pipeline is built and tested in English. If your CV and job description are in English, everything works as described. If either is in another language, the models will generally handle it - but the critique scoring, authenticity dimensions, and banned-phrase list are all calibrated against English-language norms. Output quality in other languages is not guaranteed to match what you would get in English.