11 hours ago
Berlin, GermanyMid Level
Responsibilities
- Build and own end-to-end AI features from concept through production, including document versioning and agentic chat workflows.
- Design and implement database schemas, LLM integrations and tuning, evaluations, error analysis, and user interfaces.
- Contribute to workflow harnesses, decision-trace standards, evaluation tooling, and reusable human review and approval surfaces.
- Model construction-domain ontologies and project graphs, including entity resolution across contracts, drawings, and bills of quantities.
- Extend multimodal ingestion for PDFs, scans, drawings, and 3D models at scale.
- Build read/write integrations with customer document-management and ERP systems.
- Contribute to scalable data pipelines and native BIM/DWG file processing.
Requirements
- At least 3 years of experience as a full-stack software engineer, ideally in a fast-paced early-stage B2B startup.
- Proven experience delivering production-grade applied AI features to customers.
- Proficiency with TypeScript, Python, and SQL.
- Experience building backend APIs and database schemas with Fastify, Prisma, or equivalents.
- Experience integrating and optimizing LLM APIs such as OpenAI, Azure, Gemini, or Cohere.
- Experience with AWS infrastructure such as RDS, S3, and Elastic Beanstalk, or equivalent cloud infrastructure.
- Experience building data pipelines and ETL workflows for unstructured data at scale.
- Experience with vector databases, semantic search, AI evaluation frameworks, and error analysis.
- Experience building internal platform or developer-facing tooling for other engineers.
- Experience with workflow orchestration tools such as Prefect or Airflow is a plus.
- Experience with document processing, OCR, or multimodal data ingestion is a plus.
- Experience with observability and monitoring tools such as Datadog or Langfuse is a plus.
- German language skills are helpful but not required; the team works in English.
Benefits
- Salary range is 80,000 to 120,000 EUR annually.
- On-site in Berlin, Germany, five days per week.
- Candidates must be based in Berlin or willing to relocate.
- Visa sponsorship is not available for this role.