Back to jobs

AI Engineer (Search & Matching) - fully remote within Germany Remote

Pay
Pay not listed
Work setting
Remote — work setting source
Listed location: Remote DE
Read the full posting
Employment
Permanent — employment source
Employment type in the employer’s posting: Permanent employee
From the employer’s posting

What you’ll work on

Full posting
  • You'll work closely with the people who make the decisions and present your results to them directly - your benchmark can become a production experiment within days.

From the employer’s posting
The role evolves with the mission: whatever it takes to make our search the best job search on the internet is what you'll work on next. That makes out-of-the-box thinking a feature of the job, not a nice-to-have - we expect you to bring new approaches, methods, and frameworks to the table, whether they come from commercial model providers or straight from current research. If a paper published last month suggests a better way to retrieve, rank, or evaluate, we want you to be the person who spots it, tests it, and tells us whether it holds up on our data. What stays constant is the way of working: form a hypothesis, benchmark it rigorously against the status quo, make it tangible in a prototype people can play with, ship the winner to production, and prove it in an A/B test. You'll work closely with the people who make the decisions and present your results to them directly - your benchmark can become a production experiment within days. What You'll Be Doing:

What you’ll bring

All qualifications

Core experience

  • Degree in Computer Science, AI, or a related field from a top-tier program (e.g., TUM, University of Tübingen, Saarland University, RWTH Aachen, LMU Munich, TU Berlin, KIT, University of Freiburg, University of Bonn, FAU Erlangen-Nürnberg, TU Darmstadt, University of Stuttgart, Heidelberg University, HPI (Hasso Plattner Institute, Potsdam), TU Dresden, Uni Paderborn, or equivalent)
  • Experience with latency-sensitive production ML and knowing where in a product expensive calls are acceptable
  • Communication is direct and warm, and decisions are made on evidence rather than hierarchy: benchmarks settle debates here.
  • Experience running or analyzing A/B tests on search or recommendation systems
Qualification wording
Degree in Computer Science, AI, or a related field from a top-tier program (e.g., TUM, University of Tübingen, Saarland University, RWTH Aachen, LMU Munich, TU Berlin, KIT, University of Freiburg, University of Bonn, FAU Erlangen-Nürnberg, TU Darmstadt, University of Stuttgart, Heidelberg University, HPI (Hasso Plattner Institute, Potsdam), TU Dresden, Uni Paderborn, or equivalent)
Experience with latency-sensitive production ML and knowing where in a product expensive calls are acceptable
Friendly, kind, and fast - at a sustainable pace. We move quickly because the work is focused, not because anyone burns out. Communication is direct and warm, and decisions are made on evidence rather than hierarchy: benchmarks settle debates here.
Experience running or analyzing A/B tests on search or recommendation systems
Apply at JobLeads

Posting record

Report an error

Open 3 monthsWe first saw this posting 114 days ago and it is still up. Roles open this long are often slow to fill or kept open to collect applications.

Status in our records
Live
First seen by us
Jul 30, 2026114 days ago
Recorded sightings
62
Last seen by us
Nov 21, 2026

These dates show when we found the posting. Check the employer’s website to confirm it is still accepting applications.

How this job compares

Time since we first saw it

Day 114

Up longer than most. Similar jobs usually come down between day 1 and day 15.

From 1,368 similar postings that came down. Coming down is not always a hire.

JobLeads is hiring

2 open jobs

Open jobs by the week we first saw them, last 12 weeks.

Hiring record

See how this role fits your experience

Add your resume and we’ll mark which of these tools you already use, and point out anything in the posting that could rule you out.

  • Python
Where the posting names them
- Have genuine data-science depth: you design experiments, build baselines, and reason about metrics and their failure modes before trusting a number - Are fluent in Python and comfortable owning a pipeline end to end: data, prompts, embeddings, retrieval, evaluation, and the glue in between - Don't trust a demo: you build a baseline, a dataset, and a metric before declaring victory - and you notice when your metric is measuring the wrong thing

Your next step

Complete your application on jobleads.jobs.personio.com. The employer’s form will show what is required.

Applying through Personio

  • No account needed. The 20 MB limit covers all your files together, not each one.

Already applied? Track this application

Job description

View original posting ↗

What it's all about

The Team:
JobLeads helps millions of professionals across 40+ countries find their next role, matching them against millions of live jobs. We are building what we believe will be the best job search on the internet - one that understands what a candidate is actually looking for, even when they can't put it into a keyword. Search, matching, and recommendations sit at the core of our product, and applied AI is how we are already winning there.


Your Role:
As an Applied AI Engineer, you will build the intelligence behind how candidates and jobs find each other: semantic search, resume-based matching, LLM-driven query understanding, and recommendations.

The role evolves with the mission: whatever it takes to make our search the best job search on the internet is what you'll work on next. That makes out-of-the-box thinking a feature of the job, not a nice-to-have - we expect you to bring new approaches, methods, and frameworks to the table, whether they come from commercial model providers or straight from current research. If a paper published last month suggests a better way to retrieve, rank, or evaluate, we want you to be the person who spots it, tests it, and tells us whether it holds up on our data.

What stays constant is the way of working: form a hypothesis, benchmark it rigorously against the status quo, make it tangible in a prototype people can play with, ship the winner to production, and prove it in an A/B test. You'll work closely with the people who make the decisions and present your results to them directly - your benchmark can become a production experiment within days.


What You'll Be Doing:

  • Design and improve retrieval systems over a large, messy real-world corpus at scale: embeddings, hybrid search, ranking, and re-ranking
  • Shape the design of our vector search backbone: how jobs and resumes are represented, which embedding models to use, the right vector dimensionality for each purpose, etc.
  • Build LLM pipelines for understanding queries, resumes, and jobs — extraction, enrichment, rewriting, matching - choosing the right model and technique for each step
  • Own evaluation as a scientific discipline, not a checkbox:
    • Curate benchmark datasets that capture both the failure modes you're fixing and the healthy cases you must not regress
    • Design LLM-as-judge setups you can defend - ground and calibrate the judge against human sanity checks, detect its biases (a conservative judge will punish a correct query expansion for not being literal), and iterate on the judge itself when it measures the wrong thing
    • Apply and adapt search-relevance methodology - precision@k, NDCG, pairwise preference testing - choosing metrics that resist gaming, and recognizing when a metric rewards degenerate behavior
    • Read results like a scientist: cumulative vs. independent effects, why two valid metrics can disagree, what a distribution's tail says that its mean hides
    • Quantify the full picture - quality, latency percentiles, and cost per query
  • Build interactive prototypes so product managers and stakeholders can test your approaches hands-on before anything ships
  • Partner with engineers to bring winners to production faithfully - no gap between what was benchmarked and what ships - and follow through into A/B test readouts
  • Track the fast-moving model and research landscape and re-evaluate as new models, techniques, and papers appear, keeping us on the best quality-per-euro frontier
  • See your work impact the way millions of job seekers find their next job in 42 countries
  • Be surrounded by colleagues who understand you and help you grow in your role


What You'll Need:

  • Degree in Computer Science, AI, or a related field from a top-tier program (e.g., TUM, University of Tübingen, Saarland University, RWTH Aachen, LMU Munich, TU Berlin, KIT, University of Freiburg, University of Bonn, FAU Erlangen-Nürnberg, TU Darmstadt, University of Stuttgart, Heidelberg University, HPI (Hasso Plattner Institute, Potsdam), TU Dresden, Uni Paderborn, or equivalent)
  • Have hands-on experience building applied ML or LLM systems - retrieval, RAG, agentic pipelines, or traditional search - that real users touched
  • Have genuine data-science depth: you design experiments, build baselines, and reason about metrics and their failure modes before trusting a number
  • Are fluent in Python and comfortable owning a pipeline end to end: data, prompts, embeddings, retrieval, evaluation, and the glue in between
  • Don't trust a demo: you build a baseline, a dataset, and a metric before declaring victory - and you notice when your metric is measuring the wrong thing
  • Communicate clearly with non-ML colleagues and enjoy making your findings understandable, not just correct
  • Are pragmatic and 80/20-minded: you'd rather ship a simple approach that captures most of the gain than a complex one that looks better on paper


Nice If You Have:

  • A track record of taking methods from research papers into working systems - or contributions of your own (publications, open source, competition results)
  • Experience with latency-sensitive production ML and knowing where in a product expensive calls are acceptable
  • Experience running or analyzing A/B tests on search or recommendation systems
  • Exposure to multiple model providers and cost/latency benchmarking across them


What You Can Expect On Board:

  • One of the most exciting applied-AI roles in Germany. This is a rare chance to own the intelligence behind how millions of people find their next job — at real scale, with the latest tooling, and a short path from idea to production.
  • You will never be bored - in the best sense. The problems are at the frontier of applied AI, the loop from idea to live experiment is short, and there's always a harder, more interesting question behind the one you just answered.
  • The best tools, by default. You'll work with the latest AI tooling - Claude Code and whatever comes next - so you move fast, skip the busywork, and spend your energy where your potential actually shows.
  • Friendly, kind, and fast - at a sustainable pace. We move quickly because the work is focused, not because anyone burns out. Communication is direct and warm, and decisions are made on evidence rather than hierarchy: benchmarks settle debates here.
  • Amazing colleagues. Expect good conversations and good moments - experienced product managers, engineers, and data scientists who give real feedback, get genuinely excited about a clean result, and celebrate wins together.
  • Memorable team experiences, from EU meetups to our annual company summer getaway.


We’d love to hear from graduates and PHDs of top-tier AI and Computer Science programs, particularly those with strong machine learning, LLM, or AI systems experience. This includes universities such as TUM, the University of Tübingen, Saarland University, RWTH Aachen, LMU Munich, TU Berlin, KIT, the University of Freiburg, the University of Bonn, and FAU Erlangen-Nürnberg - but outstanding candidates from other universities are equally encouraged to apply.

Demand for AI Engineer roles

New AI Engineer postings at the employers we follow are up 340% over the last 6 weeks compared with the 6 before.

Weekly new AI Engineer postings at the same employers, from the week of August 17 to the week of November 9. The fewest in a week was 78, the most 803, and the latest week had 438.
August 17Last week

Each bar is one week of new postings. The darkest bars are the last 6 weeks.

Counted from the hiring pages we’ve followed for at least five weeks, so employers we add don’t look like growth. Updated weekly.

About the role

We file this job under AI Engineer, by the work the posting describes.

AI engineers build products on top of AI models that already exist, most often large language models: they connect a model to a company's data and tools and ship the assistants, agents and search features people use.

Postings ask for strong Python or TypeScript, experience with model APIs, retrieval and agent frameworks, and ways to test whether the model's answers are good enough.

An ML engineer trains and serves models; an AI engineer usually starts from a model someone else trained and builds the product around it, though titles often mix the two.

Source details

View original posting ↗

Source notes

Source excerpts

Selected passages from the saved posting. Check the full description for conditions and exceptions.

Pay

No pay amount identified in the saved description.

Location & working pattern

Remote DE

What You'll Be Doing: - Design and improve retrieval systems over a large, messy real-world corpus at scale: embeddings, hybrid search, ranking, and re-ranking - Shape the design of our vector search backbone: how jobs and resumes are represented, which embedding models to use, the right vector dimensionality for each purpose, etc.
Work authorization

No clear work-authorization passage found. Eligibility is unconfirmed.

Find answers in the posting

AI
How answers work

AI selects complete passages from this posting. Check them for conditions and exceptions.

Uses this posting and your question. No profile needed.