Project match · with Jev
Does Michael fit your project?
Paste a job posting, upload a file or pick an example. Jev, a model by TypeSafe, checks every requirement against the CV and names the evidence. The code of this page then computes the score, traceable and adjustable.
Read the posting
Freelance: Senior Test Automation Engineer with an AI focus Utility company, customer portal, start now, 6 months, mostly remote Your tasks: - Build end-to-end test automation with Playwright and TypeScript for our customer portal - Automated API tests for the REST interfaces to billing and contract management - Integrate the test runs into Azure DevOps pipelines, including reports for the team - Use AI to generate and maintain test cases from requirements - Analyse test failures and incidents, also with the help of LLMs - Coach the development team on test strategy and the test pyramid Your profile: - At least 5 years of experience automating tests for web applications - Very good knowledge of Playwright or comparable frameworks - Experience with CI/CD, ideally Azure DevOps - Hands-on experience with LLMs in quality assurance - Load testing skills are a plus - Very good German
Read the posting
Project: AI architect for an internal agent platform Insurer, claims department, 9 months, hybrid in Frankfurt Tasks: - Design a platform on which several AI agents take over business tasks - Connect internal systems through the Model Context Protocol (MCP) - Agent-to-agent communication via A2A or comparable protocols - Retrieval over internal documents (RAG) with traceable sources - Build evaluation and tests for agent answers - Run it on Azure with monitoring and logging - Workshops with the business unit and IT for the rollout Requirements: - Proven projects with LLMs in production - Experience with agent frameworks and tool integration - Very good Python skills - Experience with Azure - Knowledge of the insurance industry is desirable
Read the posting
Test manager for an SAP S/4HANA migration Retail group, move from SAP ECC to S/4HANA, 12 months, on site in Düsseldorf Your tasks: - Own the test strategy and test planning of the migration - Steer the integration tests and the user acceptance tests with the business units - Defect management and reporting to the project management - Introduce test automation for SAP processes with Tricentis Tosca - Safeguard the data migration by reconciling the old and the new system Your profile: - Several years of experience as a test manager in SAP projects - Solid knowledge of SAP S/4HANA, ideally SD and MM - Experience with Tricentis Tosca - ISTQB certification - Experience leading test teams - Business fluent German
0 of 12000 characters
An example replays a stored measurement. Your browser sends no request for it.
Overall score
Requirements of the project
Evidence from the CV
For each requirement Jev picks the entry that shows it best.
Open points
Here Jev finds no evidence in the CV, or only related experience. That does not mean the experience is missing, only that the public CV does not show it.
Weighting
The judgements stay as they are. The weights only change the arithmetic in code, no new request is needed.
What now?
The score does not replace a conversation. If the project is interesting, write to Michael or ask his agent.
With the CV as a PDF behind it. Made in your browser.
How the score comes about
Jev writes no text. It answers small, fixed questions with a probability, a choice or a level. The code of this page composes the score from them.
- 01
Code splits
The posting is split into lines: every list item one line, a paragraph without a list into sentences. Headings drop out, at most 16 lines go out.
- 02
Jev judges
One request per line with the whole CV: is it a requirement, is it mandatory, which area does it belong to, which entry shows it and how well, from 0 to 3. One more request says how central each area is to the project.
- 03
Code computes
Coverage, must-haves and topic fit are arithmetic over these answers. Moving the weights changes the arithmetic, not the judgements.
The arithmetic
- Coverage = sum of weight × level/3, divided by the sum of the weights. A mandatory line weighs double.
- Topic fit = per area the mean coverage of its lines, weighted by how central the area is to the project.
- Overall = 70 % coverage + 30 % topic fit, rounded to 100 points. From 85 very good, from 70 good, from 50 partial.
- A line counts as shown from level 1.5, closer to "done in one role" than to "related".
Limits
- Jev only sees the public CV of this site. What is not there, it cannot show.
- The questions to Jev are in English, the texts are often German.
- Every line is asked once. A second run may differ slightly on single levels.
- The score is a machine estimate over passages of text, not a judgement about people and not a commitment.
One request to Jev, shortened
The CV entries sit in the state once, and the choice points at them by path.
POST https://api.typesafe.ai/v1/systemone
{
"model": "jev-latest",
"state": {
"requirement": "Build end-to-end test automation with Playwright and TypeScript for our customer portal",
"cv": {
"x-profile": "Freelance Quality Engineer & AI Architect. Quality Engineer and AI Architect. Quality engi …",
"s-tu-darmstadt": "Guest Lecture: AI-Assisted Software Development in Practice, TU Darmstadt (04/2026): Guest …",
"…": "…"
}
},
"questions": {
"must": {
"type": "noul",
"instructions": "Is `requirement` phrased as mandatory rather than optional or nice to have?"
},
"evidence": {
"type": "choice",
"instructions": "Which entry of `cv` best shows that the candidate meets `requirement`?",
"criteria": {
"x-profile": "`cv.x-profile`",
"s-tu-darmstadt": "`cv.s-tu-darmstadt`",
"…": "…",
"none": "No entry of `cv` shows it"
}
},
"level": {
"type": "score",
"instructions": "How well does `cv` show that the candidate has done what `requirement` asks for?",
"criteria": [
"Not at all",
"Adjacent or transferable experience",
"Done in at least one role or project",
"Done repeatedly, across several roles or projects"
]
},
"…": "…"
}
}