AI Software Engineer

United Imaging North America•Houston, TX

22h

About The Position

The AI Software Engineer designs, develops, and maintains AI-powered software that supports United Imaging’s business operations. The role combines software engineering with hands-on large language model (LLM) application development and works with business users across quality, manufacturing, finance, HR, and R&D to turn operational needs into reliable, secure, stable, and maintainable software. The position focuses on applied AI for Speed, United Imaging’s in-house AI workspace platform. Speed provides an AI assistant and agents that work across company files, collaboration tools such as Lark and Microsoft Outlook, and business systems. Capabilities include answering questions, analyzing documents and spreadsheets, running Python analyses in a secure sandbox, automating approval and reporting workflows, and supporting manufacturing-record quality reviews. The engineer works end to end, from discovering use cases with business users through building, evaluating, deploying, and measuring the impact of LLM capabilities. The role includes developing production software, LLM-powered agents and assistants, retrieval-augmented generation (RAG) and knowledge-base features, document-review workflows, and quantitative evaluation pipelines. It also involves operating commercial LLM APIs and self-hosted open-weight models while balancing quality, latency, cost, enterprise data privacy, and security. The position requires working in an environment where strong magnetic fields may be present; ferrous metals are prohibited in designated work areas, and employees must disclose relevant implants or foreign objects to their supervisor before starting work.

Requirements

Bachelor’s degree in Computer Science, Software Engineering, or a related field.
Proficiency in Python and TypeScript and experience with source code management tools such as Git.
Hands-on experience building and shipping applications on LLMs, including prompt engineering, function/tool calling, structured outputs, and streaming.
Experience with RAG, including embeddings, vector search (such as pgvector or a dedicated vector database), and retrieval-quality evaluation.
Strong Python skills, including building backend services and REST APIs, such as with FastAPI and asynchronous programming.
Experience designing quantitative evaluations for AI/ML systems, including test sets, metrics, and error analysis, and using results to guide improvements.
Complex problem-solving and critical-thinking skills, strong interest in technology, and strong IT skills.
Communication skills and professional proficiency in written and spoken English.

Nice To Haves

Master’s or PhD in Computer Science, Machine Learning, or a related field.
Experience building AI agents involving multi-step planning, tool use, sub-agents, or memory, or experience with the Model Context Protocol (MCP).
Experience serving open-weight models on premises, such as with vLLM, GPU inference, or quantization.
Experience with document AI, including OCR, layout and table extraction, and vision-language models for PDFs, spreadsheets, and scanned forms.
Full-stack experience with TypeScript, React/Next.js, PostgreSQL, and Redis.
Experience integrating enterprise platforms such as Lark or Microsoft 365 (Graph API) through OAuth and REST APIs.
Experience with medical imaging-related projects.
Passion for solving complex technical challenges with innovative ideas.

Responsibilities

Understand business operational processes and provide software solutions.
Develop flowcharts, layouts, and documentation to identify requirements and solutions.
Produce specifications, determine operational feasibility, integrate software components into functional systems, and develop software verification plans and quality assurance procedures.
Document and maintain software functionality, comply with project plans and process requirements, deploy programs, evaluate feedback, and upgrade software.
Design and build LLM-powered features for Speed, including tool-calling AI agents, multi-step workflow automation, and conversational assistants grounded in company documents and business systems.
Develop RAG and knowledge-base capabilities, including parsing PDF, Word, and Excel documents and scanned forms; chunking; embeddings; vector search; and answers that cite sources.
Apply AI to document review in regulated workflows, such as medical device manufacturing and quality records, with human-in-the-loop sign-off and full traceability.
Build AI evaluation pipelines using labeled datasets, offline replay of real runs, precision/recall and task-success metrics, and regression checks for prompt and model changes; use results to improve quality.
Integrate and operate commercial LLM APIs and self-hosted open-weight models, balancing quality, latency, and token cost through model routing, context management, and prompt caching while meeting enterprise privacy and security requirements.
Partner with business users in quality, manufacturing, finance, HR, and R&D to identify high-value AI use cases, prototype solutions, ship them to production, and measure adoption and impact.

Benefits

Medical insurance
Dental insurance
Vision insurance
Short- and long-term disability insurance
Employee assistance program
Company-paid basic life insurance
401(k) with employer match
Paid time off
Sick leave
12 paid holidays
Position may be eligible for performance-based incentive compensation, such as cash bonuses.