AI Software Engineer (LLM & Full-Stack Systems)

Pure Group

1. Generative AI Engineering

• Fine-tune, quantize, and deploy open-source LLMs (e.g., Llama-3, Mistral, Falcon, DeepSeek, Phi, etc.) for specific domains and clients
• Build and maintain multi-agent orchestration systems using frameworks like LangChain, LlamaIndex, or custom Python pipelines.
• Create and maintain retrieval-augmented generation (RAG) pipelines with vector databases (Qdrant, Pinecone, Weaviate, FAISS).
• Create architectures for different AI services, optimize prompt templates, context windows, and chain execution for efficiency and precision.
• Integrate and evaluate speech, vision, and multimodal models for production-ready use cases.

2. LLM Platform Management & Production Operations

• Deploy and maintain LLM inference servers (vLLM, Ollama, TGI, or custom microservices).
• Apply prompt caching, embedding pre-computation, and batch inference for performance optimization.
• Evaluate model performance and quality using automated feedback loops and analytics dashboards.

3. Full-Stack & Systems Development

• Architect, develop, and maintain frontend interfaces (React / Next.js) and backend services (FastAPI / Node.js).
• Design robust REST / GraphQL APIs for interaction with AI microservices.
• Integrate vector databases, relational DBs, and cloud storage in scalable data architectures.
• Contribute to both rapid prototyping (Streamlit / Gradio) and enterprise-grade platforms.

4. Collaboration & Product Development

• Work cross-functionally with AI researchers, product designers, and cloud architects to deliver complete user experiences.
• Participate in technical planning, code reviews, and performance audits.
• Document internal tools, APIs, and best practices for future developers.
• Contribute to architectural decisions for scalability, cost optimization, and compliance.