- Company Name
- Trickstr
- Job Title
- Ingénieur(e) Data & LLM (Python)
- Job Description
-
Job Title: Data & LLM Engineer (Python)
Role Summary: Design, develop, and production‑grade deploy technical components around large language models (LLM) for real‑time decision support products.
Expectations: Deliver end‑to‑end solutions, from data ingestion to API services, ensuring quality, robustness, and compliance with business objectives. Collaborate cross‑functionally with developers, data engineers, analysts, and stakeholders to iterate on product features.
Key Responsibilities
- Build and maintain data ingestion and preprocessing pipelines (cleaning, normalizing, enriching, feature engineering, scoring, aggregation).
- Implement RAG systems (search, context retrieval) and evaluation tools (hallucination detection, bias assessment, robustness, answer quality).
- Prototype in notebooks and transition models/services to production (APIs, scheduled jobs, Docker containers).
- Document code, APIs, and processes for team reuse.
- Conduct applied R&D: benchmark new models, libraries, and architectures; evaluate cost, latency, and quality trade‑offs.
- Participate in code reviews, pair programming, and knowledge transfer to non‑technical stakeholders.
- Support continuous improvement of data pipelines and model performance monitoring.
Required Skills
- Proficiency in Python (data structures, functions, virtual environments, testing).
- Experience with pandas, numpy; familiarity with FastAPI, SQLAlchemy, or similar web/DB frameworks.
- Practical knowledge of LLM APIs (OpenAI, Anthropic, etc.), prompt engineering, RAG, and light fine‑tuning.
- Strong data engineering mindset: data validation, quality metrics, reproducibility.
- Ability to measure, compare, and question model outputs.
- Team‑oriented communication: code reviews, pair programming, explaining technical concepts to non‑technical audiences.
- Bonus: Docker & deployment, vector databases, graph databases, search technologies, open‑source contributions.
Required Education & Certifications
- Graduate from an engineering school or equivalent with a background in mathematics, computer science, or related field.
- 1–3 years of industry or research experience in data engineering or AI development.
- Certifications in data engineering, machine learning, or cloud platforms are a plus.