Senior ML Data Processing Developer

lawzero · Berlin; Montreal

Beim Arbeitgeber bewerben
Gefragte Sprachen
Deutsch gute Kenntnisse
Stellenangebot auf
Englisch
Arbeitsort
Berlin; Montreal
Arbeitsmodell
Vor Ort
Veröffentlicht
17. November 2026

Kurz zusammengefasst

LawZero is a Montreal‑based non‑profit that builds safe‑by‑design AI systems, guided by research from Professor Yoshua Bengio. The organisation is looking for a Senior ML Data Processing Developer to own and evolve the core data‑asset pipeline that turns raw, web‑scale text into high‑quality training sets for its Scientist AI. The day‑to‑day work centers on partnering with the Research team to design, automate and scale pipelines that handle massive unstructured text collections. You will build and maintain processing stages such as deduplication, model‑based quality scoring, heuristic filtering, toxicity removal, PII scrubbing, metadata extraction and proprietary transformations, while ensuring full dataset versioning and provenance. The role also requires implementing compliance checks, data‑quality monitoring, guardrails and alerting, as well as creating internal tools that let researchers explore and query datasets easily. Additional duties include acquiring new text corpora, conducting coverage analyses, and working with Legal & Governance to license data sources, plus developing leakage‑detection mechanisms to prevent evaluation contamination. Candidates must hold a degree in computer science, software engineering or a related discipline and have at least five years of experience processing trillion‑token‑scale text datasets in data‑processing, ML engineering or NLP roles. Proven expertise with distributed processing frameworks such as Spark, Ray or Flink, and strong Python skills for production‑grade code are essential. Experience with pipeline orchestration tools (Airflow, Prefect, Dagster), data‑privacy implementations (PII scrubbing), content‑safety filtering (toxicity, bias) and mechanisms to avoid evaluation leakage is required. The role also expects demonstrated ability to collaborate across Research, Engineering and Legal/Governance teams. Preferred, but not mandatory, skills include experience training or fine‑tuning models for data‑quality tasks and familiarity with LLM inference optimizations (e.g., vLLM, SGLang). Knowledge of containerised deployment (Docker, Kubernetes), infrastructure‑as‑code, ML experiment‑tracking tools such as Weights and Biases, and prior work on web‑scale data acquisition or open‑source NLP tooling are considered advantageous. LawZero emphasizes an inclusive, respectful workplace and welcomes applicants who share its mission of advancing safe AI for the public good. The position is based in Montreal, offering the chance to work at the frontier of AI data engineering within a collaborative, mission‑driven environment. What the role asks for: - Degree in computer science, software engineering or related field - 5+ years handling trillion‑token unstructured text datasets - Strong Python proficiency for production data processing - Experience with distributed frameworks (Spark, Ray, Flink) - Experience with pipeline orchestration (Airflow, Prefect, Dagster) - Experience implementing PII scrubbing and content‑safety filtering - Ability to work across Research, Engineering and Legal/Governance teams - Nice‑to‑have: experience with LLM‑based data‑quality models and inference optimization.

Beim Arbeitgeber bewerben

Passt dein Lebenslauf zu dieser Stelle?

Die Überprüfung vergleicht Sprachen, Erfahrung und Fähigkeiten mit diesem Stellenangebot und kostet nichts.

Lebenslauf prüfen

Ein Lebenslauf pro Bewerbung

Die Künstliche Intelligenz bringt deinen Lebenslauf in die richtige Form und Sprache für diese Stellenanzeige, und zwar im deutschen Format.

Lebenslauf anpassen