Senior AI Developer

We need to output HTML fragment with sections: About the role? The posting includes Department Summary, Duties & Responsibilities, Qualifications (minimum and preferred), Benefits, Salary Range, Unit, Department, FLSA Exemption Status, Hours per Week, Restricted Position, ORP Eligible, Flexible Work Arrangement, University Job Title. We must not repeat job title as heading. Use

for section headings: About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule. We need to map content accordingly. We have Department Summary (maybe About the role). Duties & Responsibilities -> Responsibilities. Qualifications -> Requirements (maybe combine Minimum and Preferred). Benefits -> Benefits. Salary Range -> Pay. Schedule -> Hours per Week (40). Also maybe include Flexible Work Arrangement? Not required but could be part of Schedule? The instruction: include a section ONLY when posting has real, specific content for it. So we can include Pay (salary range), Schedule (hours per week), Benefits, Responsibilities, Requirements (Qualifications). About the role could be Department Summary. We must not include headings with filler like "commensurate with experience". We have specific salary range, so Pay section is okay. Schedule: Hours per Week: 40 (specific). So we can include Schedule. We must not include Job ID/requisition numbers as headings; we can fold minor metadata into short

or drop it. The University Job Title line: "University Job Title: 24255N - Sr IT Developer/Programmer 2 - Open". We should not repeat job title as heading, but we could include it in a short

maybe? The instruction: do NOT put Job ID / requisition numbers as headings — fold minor metadata into a short

or drop it. So we can include a

with that line? It's minor metadata; we could drop it. Safer to drop it. Also we must remove boilerplate like EEO etc. Not present. We must preserve all substantive information; do not invent or summarize away detail. So we need to include all details from Department Summary, Duties & Responsibilities (including bullet points), Qualifications (minimum and preferred), Benefits, Salary Range, Hours per Week, maybe Flexible Work Arrangement? It's specific: Fully Remote. Could be part of Schedule? But schedule section is for hours per week; we could add another section like Work Arrangement? Not in allowed headings list. The allowed headings are limited to those listed; we should not create other headings. So we could incorporate Flexible Work Arrangement into Schedule? Not ideal but maybe we can add a note in Schedule

about 40 hours per week, fully remote. Or we could add a separate

after Schedule? The instruction: Use

for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule). It says e.g., meaning those are examples; we could also use other headings? It says "Use

for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule)." Likely they expect only those headings. So we should stick to those. Thus we will have sections: About the role (Department Summary), Responsibilities (Duties & Responsibilities), Requirements (Qualifications), Benefits, Pay, Schedule. We need to use

• for any list of duties/requirements/benefits. So for Responsibilities we need to list duties as
  • . For Requirements we need to list qualifications as
    • . For Benefits we need to list benefits as
      • . For Pay we just have salary range; could be

        . For Schedule we have hours per week and maybe note fully remote; could be

        . We must preserve all substantive information: need to include all duties bullet points exactly as given. Let's extract duties: From Duties & Responsibilities paragraph: "The Senior AI Developer will be a member of the OVPRI IT team and is responsible for the design, development, and maintenance of customized artificial intelligence solutions that enhance the research mission of the University. The Developer will leverage Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) to create secure tools that will accelerate and enhance the research administration process. The incumbent must balance high-level machine learning engineering with a deep commitment to digital ethics and institutional data sovereignty." Then "Typical Duties" list: - LLM Development and Management - Custom LLM Implementation: Design and maintain LLM models that allows the research community to access models via a secure, authenticated internal interface. - RAG Pipeline Construction: Build and optimize Retrieval-Augmented Generation (RAG) pipelines that allow AI agents to "read" university-specific research administration and regulatory affairs data (grant proposals, contracts, agreements, protocols, etc) without leaking that data into public training sets. - Vector Database Management: Manage and scale vector databases to ensure high-speed semantic search across research administration and regulatory affairs documents. - New AI Documentation - Document LLM and RAG processes and configurations - Document APIs to allow applications to utilize AI technologies developed - Provide recommendations on appropriate training documents for models and fine-time parameters. - Provide detailed training documentation to users as needed. - Research AI Technologies - Research new AI technologies and provide recommendations on implementation. - Test new AI methodologies and usages and provide recommendations for integration with existing systems We need to preserve all that. We'll format as nested lists? The instruction: Use

        • for any list of duties/requirements/benefits. Could be flat list; but we can embed sublists. Probably okay. For Requirements: Minimum Qualifications and Preferred Qualifications. Minimum Qualifications bullet points: - Demonstrated ability to work in and foster an environment of respect, professionalism and civility with a population of faculty, staff, and students from all backgrounds and experiences, or a commitment to do so as a staff member at VCU. - Only applicants who do not require current or future sponsorship for employment will be considered - Bachelor’s degree in computer science, Data Science, Information Technology, or a related field. - Minimum of 2 years of experience in software development with at least 1 year of hands-on experience developing applications using LLM APIs (OpenAI, Anthropic, or open-source models); or an equivalent combination of education and professional experience. - Advanced proficiency in Python, including experience with asynchronous programming and data manipulation libraries (Pandas, vLLM, LangChain, PyTorch). - Demonstrated experience building with RAG (Retrieval-Augmented Generation) architectures and orchestration tools (like document chunking, embedding, and retrieval inferences). - Experience with at least one Vector Database (e.g. Chroma, Qdrant). - Ability to translate complex technical AI concepts for non-technical stakeholders (Faculty, Staff, and Students). - Ability to manage tasks on multiple concurrent projects. Preferred Qualifications: - Subspeciality or Degree in a field related to Artificial Intelligence. - Previous experience working within a Higher Education environment. - Experience utilizing an HPC cluster (slurm/ bash scripts and GPU usage) - Knowledge of PEFT (Parameter-Efficient Fine-Tuning) methods, specifically LoRA (Low-Rank Adaptation) - Basic ability to build rapid prototypes to demonstrate AI tools to users. - Demonstrated knowledge of data security and accessibility standards. We need to preserve all. Benefits: "All full-time university staff are eligible for VCU’s robust benefits package that includes comprehensive health benefits, paid annual and holiday leave, generous tuition benefits, retirement planning and savings options, tax-deferred annuity and cash match programs, employee discounts, well-being resources, abundant opportunities for career development and advancement, and more." We can list each benefit as
        • . Pay: Salary Range: 68,300 - 105,000. Schedule: Hours per Week: 40. Also Flexible Work Arrangement: Fully Remote. We could include both in Schedule

          : "40 hours per week, fully remote." About the role: Department Summary paragraph. We must not include headings with filler. All sections have real content. Now produce HTML fragment: start with

          About the role

          then

          ...

          . Then

          Responsibilities

          then
          with
          • items (maybe nested). Then

            Requirements

            then
            with
            • items (maybe separate Minimum and Preferred? Could combine but we need to preserve distinction. Could have two subheadings? Not allowed; only those headings. Could embed within
            • text indicating Minimum vs Preferred. But we need to preserve all detail; we can include headings inside
            • ? Not ideal. Perhaps we can have two
              sections under Requirements: one for Minimum Qualifications, one for Preferred Qualifications, each preceded by a

              or ? But we cannot add extra headings beyond

              . We could use

              Minimum Qualifications:

              then

              . That's allowed as prose. The instruction: Use

              for prose. So we can have a

              with bold text. That's fine. Thus under Requirements:

              Minimum Qualifications:

              ...
              then

              Preferred Qualifications:

              ...
              . Similarly for Benefits: we could have

              Benefits:

              ...
              but we already have heading Benefits; we can just put
              directly. For Pay:

              Pay

              Salary range: $68,300 - $105,000 per year.

              For Schedule:

              Schedule

              40 hours per week, fully remote.

              Now ensure we preserve all substantive info: include the department summary paragraph exactly. Let's copy it: "At the VCU Office of the Vice President for Research and Innovation, our mission is to foster transformative research, discovery, and innovation at Virginia Commonwealth University through excellence in service. We partner with faculty in all schools and departments as they seek funding, plan studies, establish collaborations, calculate budgets, submit grant applications, negotiate industry contracts, and secure patents and licensing agreements. Our commitment to supporting and elevating the research community at VCU has resulted in record levels of sponsored research and has placed VCU among the nation’s premier public research universities. Through our passion for discovery and willingness to take on challenges, we continue to advance the work that matters most and position the university as a national leader in research, innovation, creativity, and scholarship." We need to keep punctuation. Now duties: we need to include the introductory paragraph before Typical Duties: "The Senior AI Developer will be a member of the OVPRI IT team and is responsible for the design, development, and maintenance of customized artificial intelligence solutions that enhance the research mission of the University. The Developer will leverage Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) to create secure tools that will accelerate and enhance the research administration process. The incumbent must balance high-level machine learning engineering with a deep commitment to digital ethics and institutional data sovereignty." Then we list duties. We'll structure as:
              • LLM Development and Management
                • Custom LLM Implementation: Design and maintain LLM models that allows the research community to access models via a secure, authenticated internal interface.
                • RAG Pipeline Construction: Build and optimize Retrieval-Augmented Generation (RAG) pipelines that allow AI agents to "read" university-specific research administration and regulatory affairs data (grant proposals, contracts, agreements, protocols, etc) without leaking that data into public training sets.
                • Vector Database Management: Manage and scale vector databases to ensure high-speed semantic search across research administration and regulatory affairs documents.
              • New AI Documentation