Generative AI Engineer

SocialAngel

About the Role

This role involves building AI agents, LLM-powered applications, Retrieval-Augmented Generation (RAG) systems, and healthcare workflow automation. The ideal candidate will have practical experience in deploying AI solutions from concept to production, working with large language models, Python-based backend services, APIs, databases, and cloud infrastructure. This position requires an engineer capable of independently implementing AI features and collaborating with backend, frontend, and product teams.

Key Responsibilities

AI Agents & Agentic Workflows

• Design, develop, and deploy AI Agents for healthcare and administrative workflows.
• Build autonomous workflows for information retrieval, tool usage, multi-step task execution, and secure application API interaction.
• Implement agent orchestration, structured outputs, memory, state management, tool calling, and human approval workflows.
• Develop reusable AI Agent components for multiple healthcare use cases.
• Implement error handling, retries, permission controls, execution limits, and audit trails.

Generative AI & LLM Integration

• Integrate and optimize Large Language Models such as OpenAI, Anthropic, AWS Bedrock, or equivalent platforms.
• Develop prompt pipelines, structured outputs, context management, and model routing strategies.
• Optimize AI applications for accuracy, latency, scalability, and operational cost.
• Build reusable AI services and evaluation frameworks for consistent model performance.

AI Medical Scribe & Clinical Documentation

• Develop AI-powered medical scribing solutions to convert doctor-patient conversations into structured clinical documentation.
• Integrate speech-to-text, transcription, and speaker diarisation technologies.
• Generate draft SOAP notes, patient summaries, and structured clinical records for physician review.
• Extract symptoms, medications, allergies, diagnoses, medical history, assessments, and follow-up recommendations.
• Continuously improve documentation quality, completeness, and transcription accuracy.

Retrieval-Augmented Generation (RAG)

• Design and implement RAG pipelines using embeddings, vector databases, semantic search, and metadata filtering.
• Build intelligent search capabilities for patient records and healthcare knowledge.
• Develop context-aware patient summaries and source-grounded AI responses.
• Continuously evaluate retrieval quality and reduce hallucinations.

Patient Intelligence & Healthcare Automation

• Build AI-powered Patient 360 summaries using authorized patient information.
• Develop AI assistants for preparing referrals, patient instructions, drafting documentation, and clinical notes.
• Automate repetitive healthcare workflows using intelligent AI Agents.
• Integrate AI capabilities into existing clinical and patient management workflows, ensuring appropriate human oversight.

Backend Development & System Integration

• Build scalable AI backend services using Python and FastAPI.
• Develop secure REST APIs and integrate AI services with existing healthcare applications.
• Work with PostgreSQL, asynchronous processing, background workers, and event-driven architectures.
• Design reusable and scalable backend services following modern software engineering practices.

AI Evaluation, Monitoring & Optimization

• Develop automated evaluation pipelines for AI Agents, RAG systems, and generated clinical documentation.
• Monitor response quality, latency, failures, token consumption, and infrastructure costs.
• Implement observability using tools like Langfuse, OpenTelemetry, or equivalent platforms.
• Continuously improve AI quality using production metrics, testing, and user feedback.

Security & Healthcare Data Privacy

• Develop AI systems following secure software engineering practices.
• Implement secure API integrations, tenant isolation, audit logging, and least-privilege access.
• Protect AI systems against prompt injection, unsafe tool execution, and sensitive data exposure.
• Support HIPAA-aligned engineering practices and healthcare data privacy requirements.

Requirements & Tech Stack

Required Skills

• Strong proficiency in Python.
• Experience building backend applications using FastAPI, Flask, or similar frameworks.
• Hands-on experience with OpenAI, Anthropic, AWS Bedrock, or equivalent LLM platforms.
• Experience building AI Agents using LangChain, LangGraph, LlamaIndex, or similar frameworks.
• Strong understanding of RAG architectures, embeddings, vector databases, and semantic search.
• Experience with SQL, PostgreSQL, REST APIs, and backend integrations.
• Familiarity with Docker, Git, CI/CD, and AWS or similar cloud platforms.
• Understanding of prompt engineering, structured outputs, model evaluation, and AI optimization.
• Experience building and deploying production-grade AI applications.

Preferred Qualifications

• Experience working on Healthcare AI, Medical NLP, Clinical Documentation, or Medical Scribing solutions.
• Familiarity with EHR/EMR systems and healthcare workflows.
• Knowledge of FHIR, HL7, or SMART on FHIR.
• Experience with speech-to-text, transcription, or speaker diarisation technologies.
• Familiarity with vector search technologies such as Vectara, OpenSearch, Pinecone, or Weaviate.
• Exposure to Next.js or TypeScript is an added advantage.

Candidates with exceptional AI engineering expertise and strong production experience are encouraged to apply even without prior healthcare domain experience.