Data Engineer | Data Scientist — Unstructured Data & AI Pipelines
Neurons Lab · Remote · Poland · Portugal · Cyprus · Serbia · Hungary · Spain · Malta · Georgia · Italy · Armenia · Slovakia · Latvia · Lithuania · Greece · Albania · Czech Republic · Moldova · Ukraine · Bulgaria · Estonia · Macedonia · Romania ·
COMPENSATION
€•• – ••k
Listed on the employer's job board (Ashby)
- Part-time (20-30h/week)
- Fully remote
- Python + SQL
- Unstructured data focus
- Start ASAP
Join Neurons Lab to engineer the data foundation for a private investment group's AI initiative. You will design systems to capture, clean, and connect unstructured data from sources like Slack, email, and board documents. This part-time engagement involves building robust pipelines for entity resolution and vector storage.
What you'll do
- Ingest data from calls, emails, Slack, and documents with speaker attribution
- Parse and deduplicate historical archives including PDFs and spreadsheets
- Implement identity resolution across different communication platforms
- Build chunking and embedding pipelines for vector and graph stores
- Develop incremental sync mechanisms to avoid full re-crawls
- Attach access scope and provenance metadata to every ingested record
- Apply PII detection, redaction, and retention logic for security compliance
- Orchestrate monitored pipelines using Airflow or Step Functions
- Optimize cost and latency through batching and storage tiering
- Create runbooks for client team handover and ongoing operation
What you'll need
- At least 4 years of data engineering experience with unstructured data
- Strong proficiency in Python and solid SQL skills
- Experience parsing transcripts, mail, chat, and documents
- Knowledge of embedding infrastructure and vector stores like pgvector
- Ability to integrate APIs such as Google Workspace, Slack, and CRM
- Expertise in entity resolution without clean shared keys
- Experience with orchestration tools like Airflow or Step Functions
- Familiarity with AWS or GCP data stacks in private deployments
- Practical knowledge of PII detection, encryption, and retention
- Clear written English and ability to work asynchronously
Nice to have
- Experience building pipelines that feed LLM or retrieval systems
- Understanding of GDPR for employee-generated data and EU residency
- Knowledge of data lineage and audit patterns for AI systems
- Awareness of financial services security and cost expectations
Stack
- Python
- SQL
- Airflow
- Step Functions
- AWS
- GCP
- pgvector
- OpenSearch
- Pinecone
- Google Workspace
- M365
- Slack
How you'll work
- Part-time engagement, 20 to 30 hours per week
- Fully remote position
- Immediate start required
- Project duration of approximately eight to ten sprints
- Workload front-weighted to the first four or five sprints