Job Summary (List Format):

Job Requirements:
Hands-on Python experience, including APIs, async programming, and robust engineering fundamentals.
1–3 years of experience with GenAI or LLM applications.
Practical experience with LangChain or LangGraph for building chains, agents, and stateful LLM workflows.
Experience with cloud AI platforms, specifically Google Vertex AI (Models, Vector Search, Embeddings, Pipelines) and Azure OpenAI (GPT-4, GPT-4 Turbo, Assistants API, embeddings).
Solid knowledge of RAG pipelines, multi-step reasoning agents, tools/function calling, and prompt engineering.
Experience with vector databases such as PGVector or MongoDB.
Understanding of end-to-end GenAI workflows: ingestion, chunking, embedding, retrieval, and inference.
Proven experience building APIs with FastAPI or Flask.
Knowledge of secure AI deployment patterns, access controls, and AI safety guidelines.
Understanding of embeddings, transformer models, and fine-tuning methods.
Experience evaluating LLM outputs using BLEU, ROUGE, RAGAS, or LLM-as-judge methods.




Familiarity with CI/CD, Docker, containerization, and cloud-native deployments.
Exposure to MLOps tools such as Prefect, ADF, Airflow, or prompt testing frameworks.
Robust debugging, analytical thinking, and ownership mindset.

Key Responsibilities:
Design and develop GenAI and LLM-powered backend services using Python.
Build RAG pipelines including ingestion, chunking, embedding, retrieval, and inference layers.
Implement multi-step agent workflows, tool-calling logic, and structured reasoning flows using LangChain or LangGraph.
Integrate with Vertex AI or Azure OpenAI for model serving, vector search, and embeddings.
Integrate vector databases (PGVector, MongoDB) for semantic retrieval.
Develop APIs using FastAPI or Flask to expose LLM or GenAI capabilities as internal services.
Apply secure deployment patterns, ensuring responsible use of LLMs and APIs.
Write unit tests, eva

📌 Ai Engineer Reg Veerapur (India)
🏢 Sparix Global
📍 India