CareerPlanGet AI match score →

Software Engineer – Full Stack AI - US-Remote

Software Engineer – Full Stack AI - US-Remote🌐 Remote💼 Full-time💰 $130,000–$170,000🗓 2026-06-24

Core

Designing AI architecture standards and shipping production AI features (LLM integrations, RAG, vector search) for a healthcare communications platform.

Role type

Senior Full Stack AI Engineer (Builder)

Builds

Production AI features, LLM integrations, RAG systems, and vector search pipelines for a SaaS platform.

Domain

Healthcare technology / Generative AI / Cloud-native systems

Deliverable

production ML models | product features

Required skills

Python backend development, LLM API integration, RAG system design, vector database management, cloud-native architecture, latency optimization, cost control, security/PHI handling

Preferred skills

AWS experience, regulated environment experience, product-minded systems thinking

Technologies

OpenAI, Anthropic, AWS Bedrock, vector databases

Responsibilities

Design AI integration patterns and architecture standards, integrate LLM APIs into production features, build RAG systems and embedding pipelines, optimize latency and inference cost, establish model evaluation and observability processes, define cost controls and security constraints, evaluate use cases for ROI, advise leadership on build vs. buy decisions.

Seniority

Senior, hands-on IC

Rewrite
## About the role PerfectServe is a leading healthcare communications platform that connects providers, patients, and care teams to improve clinical collaboration and patient outcomes. We are seeking a Full Stack AI Engineer to join our R&D organization and play a central role in bringing AI capabilities to production across the platform. This is a core, high-impact role responsible for defining AI architecture standards, shipping production AI features, establishing governance and evaluation frameworks, and partnering directly with R&D leadership to shape the AI roadmap. ## Essential Functions - Design AI integration patterns and architecture standards across the SaaS platform - Integrate LLM APIs (OpenAI, Anthropic, AWS Bedrock) into production features - Build RAG systems, vector search, and embedding pipelines - Optimize latency, inference cost, and hallucination guardrails - Establish model evaluation, benchmarking, and observability processes - Define cost controls, usage monitoring, and security/PHI handling constraints - Evaluate use cases with Product for ROI and feasibility - Advise leadership on build vs. buy decisions for AI capabilities ## Qualifications - 5+ years software engineering experience with strong Python backend skills - Production experience with LLM APIs, RAG systems, embeddings, and vector databases - Cloud-native architecture experience, preferably AWS - Experience in regulated or security-conscious environments - Product-minded systems thinker, comfortable with ambiguity and executive communication This is not a research role. This is a builder role. ## Why Join PerfectServe? At PerfectServe, we are transforming healthcare communication and collaboration to help clinicians deliver better care. You'll work with a dedicated and mission-driven team in an environment that values growth, transparency, and innovation. We offer a salary range of $130,000-170,000 USD per year, with compensation tailored to your background, strengths, and potential to grow within the team. The salary range listed for this role reflects our commitment to pay transparency and is based on market data, internal equity, and the scope of responsibilities. compensation will be determined by a combination of factors, including the candidate's experience, skills, and the specific team or product area they support. We regularly review compensation across the company to ensure fairness and consistency. If you are a current employee and have questions about how your compensation aligns with our ranges, we encourage you to speak with your manager or People Operations.
Sourced via wellfound · Listed on CareerPlan, which tracks 70,000+ jobs from 20+ sources.
Apply on Wellfound ↗