Développe des outils d'intelligence artificielle pour la simulation et la conception de produits industriels.
AI Engineer: Evals
PhysicsX recrute un AI Engineer spécialisé en évaluation d’agents et de produits fondés sur les LLM. Vous concevrez des frameworks d’évaluation, des jeux de données et des benchmarks, et relierez les tests hors ligne aux données de production. Le poste est basé à Londres et implique une collaboration avec les équipes produit, recherche, ingénierie et clientèle.
3× plus de chances d’entretien avec un cold email
Voir l’offre originaleRepères sur PhysicsX
- Domaine officiel
- physicsx.ai
- Offres ouvertes
- 16
Détails de l’offre
La description complète publiée par PhysicsX.
About us
Re-architecting Engineering for the Age of Intelligence PhysicsX is the physics AI company for industrials. The company’s mission is to accelerate hardware innovation by overhauling what industrial engineering and manufacturing look like today. PhysicsX is building a new simulation software stack to deliver deep physics AI enablement across the entire engineering lifecycle. The company partners with leading organisations in aerospace & defence, automotive, semiconductors, materials, and energy & renewables, supporting them on some of their most critical and complex challenges.
PhysicsX is headquartered in the United Kingdom, with offices in London, New York, and Singapore and an expanding presence in the Bay Area. The role You will build the evaluation discipline that lets teams design, deploy and improve trustworthy agents in production. You will report into the overall engineering lead for our Agentic strategy. This is a hands-on engineering role with an organisation-wide mandate. Product, delivery and customer teams own the evals for the agents they ship. You will give them the shared infrastructure, frameworks, practices and coaching to do that well.
You will also own cross-product evaluation and develop public benchmarks for agents doing real engineering work.
What you will do
- Lead the technical strategy and implementation of how PhysicsX evaluates agents.
- Build reusable evaluation frameworks, reference implementations and golden paths on top of our existing tooling.
- Help product, delivery and customer teams establish robust eval suites, then turn what works into repeatable guidance.
- Develop datasets and evaluators that combine deterministic checks, domain-expert judgement, model-based grading and human review.
- Connect offline evaluation, regression testing and production telemetry so teams can detect drift and learn from real use.
- Define cross-product scorecards and shared benchmarks without taking ownership away from the teams shipping each agent.
- Create and publish credible open benchmarks for agents performing real engineering and manufacturing workflows.
- Work with Research and Post-Training on model and agent evaluation. Own cross-product, production-agent and public benchmark integrity; share approval when results support product or external claims.
- Build institutional muscle through reviews, examples, training and direct collaboration, making rigorous evaluation part of how teams think rather than a final release step.
- Help determine how this capability should grow and, over time, shape the team around it.
What you bring
Where you need to be strongest
- You have built and operated evaluation systems for agents or LLM-powered products in production.
- You can turn an ambiguous product goal into datasets, evaluators, experiments and decision criteria that teams trust.
- You understand the failure modes of probabilistic systems and use appropriate statistical judgement rather than treating benchmark scores as exact.
- You know when to use deterministic evaluators, domain experts, human review and LLM-as-a-judge, including the limitations of each.
- You are a strong engineer who builds usable frameworks and production-quality tooling, not only analyses or notebooks.
- You have helped multiple teams adopt a technical practice through hands-on delivery, clear standards and pragmatic coaching.
- You can work directly with customers and domain experts, learn what correctness means in their workflows and encode it without losing important nuance. Foundations
- Evaluation of multi-step agents, tool calls, structured outputs and long-running workflows.
- Trace-based observability and OpenTelemetry concepts.
- Dataset design, sampling, experiment design, regression detection and reproducibility.
- CI and release gates for systems whose behaviour is non-deterministic.
- Clear technical writing and communication across product, research, engineering, delivery and customer teams.
- Product judgement: knowing which measurements should drive a decision and which are merely interesting.
Nice to have
- Experience : with Pydantic Evals, Pydantic Logfire or comparable code-first evaluation and observability tools.
- Experience : evaluating agents in engineering, simulation, design, manufacturing or other expert domains.
- Experience : creating public benchmarks or open-source evaluation tooling. - Familiarity with post-training, synthetic data, reward signals or model selection.
- Experience : in regulated, customer-operated or air-gapped environments.
We value diversity and are committed to equal employment opportunity regardless of sex, race, religion, ethnicity, nationality, disability, age, sexual orientation or gender identity. We strongly encourage individuals from groups traditionally underrepresented in tech to apply. To help make a change, we sponsor bright women from disadvantaged backgrounds through their university degrees in science and mathematics.
We collect diversity and inclusion data solely for the purpose of monitoring the effectiveness of our equal opportunities policies and ensuring compliance with employment and equality legislation. This information is confidential, used only in aggregate form, and will not influence the outcome of your application.
Prérequis
- Avoir conçu et exploité des systèmes d’évaluation pour des agents ou produits LLM en production
- Maîtriser les évaluateurs déterministes, le jugement d’experts, la revue humaine et l’évaluation par LLM
- Expérience de l’évaluation d’agents à plusieurs étapes, des appels d’outils et des workflows longs
- Compétences en conception de jeux de données, échantillonnage, expérimentation et reproductibilité
- Expérience des tests de régression, de la détection de dérive et des processus CI et de mise en production
Candidater à cette offre
Comment candidater ?
L’offre est publiée sur le site de PhysicsX. Ouvrez l’offre originale ↗ pour y déposer votre candidature. Screeningpass ne remplit pas de formulaire à votre place.
À qui écrire chez PhysicsX pour ce poste ?
Lancez une recherche de contact depuis cette page. Screeningpass vous propose jusqu’à trois personnes liées au poste. Vous choisissez celle à qui écrire, puis Screeningpass cherche son email professionnel et prépare un message. Vous le relisez et l’envoyez vous-même avec Gmail. Une adresse email n’est pas toujours disponible, et une réponse n’est jamais garantie.
Combien coûte la recherche d’un contact ?
Une recherche coûte 1 crédit. Le crédit vous est rendu si la recherche échoue. L’inscription offre 3 crédits, soit 3 recherches.
À quoi sert Screeningpass ?
Screeningpass rassemble des offres de stage, d’alternance et d’emploi publiées par les entreprises. Pour une offre, Screeningpass trouve une personne à contacter dans l’entreprise, cherche son email professionnel et prépare un message à partir de votre CV. Voir le fonctionnement en détail.