Data Engineer bei Johnson Lambert
Johnson Lambert · Park Ridge, Vereinigte Staaten Von Amerika · Onsite
- Professional
- Optionales Büro in Park Ridge
What You’ll Do
- Own the modern data foundation on AWS: Design secure, scalable, and cost‑aware lake/lakehouse patterns using open, interoperable formats and layered architecture (raw → standardized → curated/analytics‑ready).
- Build dependable batch pipelines: Implement ingestion, transformation, validation, and orchestration to move data from source systems to governed, analytics‑ready datasets with clear SLAs/SLOs.
- Translate messy files into trusted data: Create robust, repeatable processes to extract and normalize data from Excel (multi‑sheet, merged cells, header variations, hidden rows, cross‑tab layouts) and PDF documents (including OCR and table extraction), mapping to standardized schemas.
- Integrate key SaaS sources: Ingest data via APIs/exports from business apps—Salesforce, Slack, Tableau (and similar)—and keep them in sync on reliable schedules.
- Structure data for AI/ML accessibility: Prepare datasets for analytics, ML, and LLM workloads—e.g., semantic/feature layers, curated text corpora, and vector indexes/databases for LLM retrieval (RAG), with appropriate metadata and access controls.
- Model for the business: Implement pragmatic dimensional/lakehouse models aligned to how our audit, tax, and advisory teams work—especially across insurance, nonprofit, and employee benefit plan domains.
- Raise data quality & trust: Embed tests and contracts, schema checks, and observability; maintain lineage, documentation, and data dictionaries that non‑engineers can use.
- Harden security & governance: Apply AWS identity, access controls, encryption, classification/tagging, and right‑sized governance appropriate for client‑serving environments—explicitly protecting client data used in AI/ML contexts.
- Automate and templatize: Use infrastructure‑as‑code and CI/CD to make environments reproducible; publish templates/patterns that teammates can reuse without deep data engineering expertise.
- Enable and mentor: Partner with analysts/automation engineers; run reviews, workshops, and coaching to uplevel the team and make data self‑service where practical.
Required Qualifications
- 5–7 years in progressively complex data engineering/data architecture roles.
- Strong experience building on AWS (storage, compute/serverless, identity, orchestration, monitoring) and operating secure, production data workloads.
- Proven success designing and implementing modern lake/lakehouse architectures using open, interoperable approaches (transactional tables, partitioning, governance, performance optimization).
- Expert data wrangling in Python and SQL for structured and semi‑structured data (CSV, JSON, Excel) and practical experience with PDF extraction (OCR, layout detection, table parsing).
- Hands-on experience building and deploying data infrastructure using infrastructure-as-code (e.g., Terraform, AWS CDK), CI/CD practices, and modern data testing/observability tooling.
- Practical experience implementing data governance solutions for cataloging, lineage, and documentation suitable for sensitive, client-service environments.
- Experience with ETL/ELT tools (e.g., Airflow, Spark) and data platforms such as Databricks or Snowflake; we prioritize open approaches and thoughtful tool selection.
- Ability to ingest from SaaS apps (e.g., Salesforce, Slack, Tableau) via APIs/exports and normalize these feeds into curated datasets.
- Comfortable as a player‑coach and first‑of‑its‑kind hire: setting standards, making build‑vs‑buy decisions, and delivering under ambiguity.
- Excellent communication skills to translate business requirements into clear technical plans, and vice versa.
- Bachelor’s in Computer Science or related field preferred.
- AWS certifications a plus.
Nice to Have
- Familiarity with insurance/nonprofit/EBP data (e.g., policy, claims, loss registers; donor/grant; plan/participant).
- Big data technologies (e.g., Hadoop, Kafka)—even though our current workloads are batch and not near real‑time.
- Experience with LLM‑assisted extraction or classification for document normalization (with governance/guardrails).
How You’ll Succeed (Outcomes & Measures)
- First 90 days
- Stand up or harden a secure AWS baseline and initial lake/lakehouse layout with CI/CD.
- Deliver a production batch pipeline converting one high‑value Excel/PDF process into a standardized, validated dataset with documentation and lineage.
- By 6 months
- Operationalize 2–3 priority SaaS integrations (e.g., Salesforce, Slack, Tableau) feeding curated layers on a dependable schedule.
- Reduce manual prep for target stakeholders by 30–50% through standardized schemas and self‑service access.
- By 12 months
- Publish reusable ingestion and document‑processing templates; establish data quality SLAs/SLOs adopted by multiple teams.
- Demonstrate measurable improvements in reliability, freshness, and adoption across analytics use cases.
How We Work
Our culture prizes agility, respect, and trust. We iterate in short cycles, document what we build, and keep stakeholders close. We choose modern, open, and maintainable solutions and believe governance should enable—not hinder—delivery. For more information on our benefits please visit https://www.johnsonlambert.com/careers/why-jl/
Equity note: Research suggests that women and Black, Indigenous, and other persons of color are less likely than men or White job seekers to apply for positions unless they are confident they meet 100% of the qualifications. We strongly encourage interested individuals to apply, and allow us to evaluate the knowledge, skills, and abilities you demonstrate, using an internal equity lens.
Johnson Lambert prides itself for the hands-on approach and relationships we build with future employees, employees, and clients. We believe each application is the potential for a future relationship with JL. Therefore, a member of our HR team personally reviews all applications submitted.
Jetzt bewerben