Key Responsibilities
- RAG Architecture: Design and optimize Retrieval-Augmented Generation pipelines, ensuring high-quality context retrieval through advanced embedding techniques and vector databases.
- Rapid Prototyping: Drive the innovation cycle by building high-fidelity Proof of Concepts (PoCs) for new AI-powered features.
- API & Backend Systems: Develop high-performance APIs and microservices (using Python/FastAPI or Node.js) to manage AI inference workflows and data processing.
- LLM Orchestration: Integrate and fine-tune interactions with various LLM providers (OpenAI, Anthropic, or Open Source via HuggingFace) using orchestration frameworks.
Technical Requirements
- AI & LLM Specialization:
- Frameworks: Proficiency in LangChain, LlamaIndex, or Haystack.
- Vector Databases: Hands-on experience with Pinecone, Milvus, Weaviate, or ChromaDB.
- Models: Deep understanding of prompting, context window management, and model selection across OpenAI, Claude, and Llama 3.
- Concepts:
Solid grasp of semantic search, tokenization, and embedding models.
Full-Stack Engineering
- Backend: 3+ years of experience in Python (FastAPI/Flask) or Node.js. Knowledge of.NET is a solid plus.
- Frontend: Expertise in React.js or Next.js for building responsive, AI-driven user interfaces.
- Cloud & DevOps: Familiarity with Docker and deployment on cloud platforms (AWS, Azure, or GCP).
- Databases: Experience with both SQL (PostgreSQL) and NoSQL environments.
Preferred Qualifications
- Proven track record of building and deploying AI side projects or commercial PoCs.
- Experience with Agentic Workflows (e.g., CrewAI or AutoGen).
- Strong understanding of AI evaluation metrics and monitoring.