LLM / GENAI Prototyping Specialist

Atlas; New City (NY); Remote; USD 100,000 - 130,000; Full time

Job summary

A leading technology firm is seeking a professional to design and scale LLM-powered assistants for a Life Sciences Cloud transformation program. This role focuses on creating AI solutions, collaborating with stakeholders, and improving end-user workflows. Candidates should have 2-3+ years of experience with GPT-class models, prompting strategies, and RAG systems. Strong communication skills are essential, along with the ability to work independently in a fast-paced environment. This is a remote-first position with proximity to NYC preferred.

Qualifications

• 2-3+ years of experience with GPT-class models or equivalent LLMs.
• Strong expertise in prompt engineering and system prompt design.
• Experience with unstructured content and retrieval-augmented generation (RAG) systems.

Responsibilities

• Design and prototype LLM-powered assistants for training and operational tasks.
• Convert ambiguous requests into clear conversational flows.
• Collaborate with teams to improve accuracy and design effective AI solutions.

Skills

Hands-on experience with GPT-class models

Prompt engineering

Building RAG systems

Communication with non-technical stakeholders

Tools

Salesforce

AWS

Job description

You will help stand up and scale early-stage LLM and agent-based solutions as part of a Life Sciences Cloud transformation program. This role focuses on rapid prototyping and maturation of AI-powered assistants that support internal users through training, hypercare, and operational enablement—particularly over large volumes of unstructured content.

You will translate ambiguous business needs into concrete LLM behaviors, design retrieval-augmented generation (RAG) solutions, and collaborate closely with training, product, field, and engineering stakeholders to deliver practical, high-impact AI capabilities.

This is a contract role, remote-first, with NYC or Collegeville proximity as a nice-to-have.

Job Responsibilities

• Design, prototype, and iterate on LLM-powered assistants for training, hypercare, and operational enablement use cases
• Convert loosely defined requests (e.g., "how-do-I assistant" or knowledge companion) into clear conversational flows, system prompts, and grounding strategies
• Build and tune RAG pipelines over unstructured document sets, including:
  • Metadata and tagging design
  • Ensure retrieval logic, prompt design, and response behavior are tuned holistically
• Take prototypes from proof-of-concept to more robust, enterprise-ready solutions by:
  • Designing evaluation datasets and test cases
  • Iterating to improve accuracy, consistency, and latency
• Partner with training, product, and field stakeholders to operationalize content and prioritize high-value use cases
• Collaborate effectively with platform and data engineers within an enterprise application ecosystem
• Focus on pragmatic, "base-hit" AI use cases that measurably improve day-to-day workflows for end users

Qualifications

• 2–3+ years of hands-on experience with GPT-class models or equivalent LLMs
• Strong expertise in prompt engineering, system prompt design, and grounding strategies
• Practical experience building retrieval-augmented generation (RAG) systems over unstructured content
• Clear understanding of how chunking, retrieval, and prompting interact as a system
• Experience embedding LLM capabilities into enterprise applications (e.g., Salesforce, ServiceNow, Dynamics, SAP, or custom internal platforms)
• Experience working in or adjacent to regulated environments (life sciences, healthcare, or financial services preferred)
• Ability to operate independently in a fast-moving, prototype-driven environment
• Strong communication skills and comfort working directly with non-technical stakeholders

Nice to Have

• Familiarity with Salesforce AI capabilities, including AgentForce or Einstein
• Prior work on training assistants, hypercare tools, or knowledge-based AI companions
• Familiarity with AWS fundamentals (e.g., S3, Lambda) and data pipelines
• Life sciences, pharma, or other highly regulated industry experience