InPost

Staff Data Engineer

Brak informacji o wynagrodzeniu

SeniorHybryda·WarszawaFull-time·B2B

#476189·Dodano 8 dni temu·0

Źródło: justjoin.it

Aplikuj teraz

Tech Stack / Keywords

Databricks Python Pandas Spark PySpark

Podsumowanie oferty

Design and own the feature-store architecture for credit decisioning, including data pipelines and agentic workflows. Set reusable engineering standards for Spark, Databricks and Python across the data engineering team.

Wymagania

8+ years of data engineering experience at scale and 5+ years with Spark/Databricks or equivalent lakehouse architecture.
Expert in SQL, data modelling and Python (7+ years), plus performance and cost engineering.
3+ years of experience with agentic coding tools and loop engineering, including context and tool design, evaluations and guardrails.
Deep knowledge of LLM behaviour and trade-offs; curious about advanced agentic harness, loop and graph engineering.
Experience setting standards through influence and mentoring without authority.
Strong in data quality, observability, lineage and GDPR-aware governance.
Fluent in English (B2+).

Bonus points for:

Feature-store or ML data serving in production.
Payments, fintech or lending data experience.
Streaming (Kafka/Structured Streaming) and CDC.
DBT governance experience.
CCD2/DORA awareness.

Obowiązki

The mission:

Design and own the feature-store architecture that credit decisioning will depend on.
Design schema, point-in-time correctness and freshness SLAs to be reliable and cheap to run.
Set pipeline patterns, modelling standards and lineage approaches through reference implementations.
Lead the highest-complexity, highest-risk data initiatives, including cross-domain feature reuse across regulated entities.
Design the agent loop, including context, tools, evaluations and guardrails, for reliable, observable and cost-controlled agentic engineering.
Own the architecture and inner workings of agentic harnesses and loop/graph agent workflows bound to the feature store, including feature engineering, point-in-time validation, drift and data-quality observability.
Set reference patterns for others to reuse.

InPost

68 aktywnych ofert

Zobacz wszystkie oferty

Aplikuj teraz