AI Engineer

Job description

Beghou brings over three decades of experience helping life sciences companies optimize their commercialization through strategic insight, advanced analytics, and technology. From developing go-to-market strategies and building foundational data analytics infrastructures to leveraging artificial intelligence to improve customer insights and engagement, Beghou helps life sciences companies maximize performance across their portfolios. Beghou also deploys proprietary and third-party technology solutions to help companies forecast performance, design territories, manage customer data, organize, and report on medical and commercial data, and more. Headquartered in Evanston, Illinois, we have 10 global offices. *Our mission is to bring together analytical minds and innovative technology to help life sciences companies navigate the complexity of health care and improve patient outcomes.* ## What you’ll do - Build and ship AI features for document understanding, structured extraction, and review workflows. - Develop reliable pipelines that combine language models with application logic, validation, and human review. - Design evaluations using representative examples, labeled data, and clear quality measures. - Improve output quality through prompting, retrieval, model selection, and structured generation. - Build safeguards for uncertain, incomplete, or conflicting inputs. - Partner with product, design, and domain experts to turn workflow needs into useful AI capabilities. - Work with engineers to integrate AI services into APIs, background jobs, and customer-facing applications. - Monitor quality, latency, and cost in production, and use the results to guide improvements. - Document system behavior, limitations, and operational considerations. ### What you’ll bring - 5–8 years of engineering experience, including hands-on work building and shipping applied AI or machine learning products. - Experience using language models for practical product workflows. - Strong Python skills and experience building production services or data pipelines. - Experience evaluating model outputs with measurable criteria, not just anecdotal review. - Familiarity with structured outputs, tool use, retrieval, and validation techniques. - Good judgment about model uncertainty and when a workflow should require human review. - Ability to work with ambiguous problems and explain technical trade-offs clearly. - Strong software engineering fundamentals, including testing, observability, and maintainable design. ### Helpful experience - Document AI, information extraction, OCR, or converting unstructured business documents into structured data. - LLM evaluation, dataset curation, human-in-the-loop systems, and regression testing. - FastAPI, PostgreSQL, cloud infrastructure, and background processing. - Familiarity with privacy, data handling, and production monitoring for AI features.