Python Backend Engineer

Vcitysoft Solutions; Pune District; On-site; INR 1,800,000 - 3,000,000; Full time

Job summary

Vcitysoft Solutions in Pune is seeking a GenAI Engineer to design and deploy production-grade GenAI applications using LangChain, LangGraph, and CrewAI. You will build end-to-end RAG pipelines and optimize model performance for real-world enterprise use cases.

You will collaborate with data teams to fine-tune LLMs (GPT-4, Claude, Llama) and implement multi-agent orchestration to automate workflows, while tracking latency and cost metrics.

Qualifications

• 2+ years of production experience with LLMs, prompting, or NLP in live apps.
• Strong Python skills, especially FastAPI, asyncio, and data pipelines.
• Experience with an agent framework: LangChain, LangGraph, CrewAI, or AutoGen.
• Good understanding of embedding models and vector databases for RAG.

Responsibilities

• Design and build production-grade GenAI applications using LangChain, LangGraph, and CrewAI
• Implement RAG pipelines with vector databases for enterprise knowledge systems
• Fine-tune and deploy LLMs for domain-specific use cases
• Build multi-agent orchestration systems for workflow automation
• Develop evaluation metrics to benchmark model performance, latency, and cost

Skills

LLMs in prod

Python

Agent frameworks

Embeddings & vector search

MLOps / PyTorch

Job description

• Design and build production-grade GenAI applications using LangChain, LangGraph, and CrewAI
• Implement RAG pipelines with vector databases (Pinecone, Weaviate, ChromaDB) for enterprise knowledge systems
• Fine-tune and deploy LLMs (GPT-4, Claude, Llama) for domain-specific use cases
• Build multi-agent orchestration systems for complex workflow automation
• Develop evaluation frameworks to benchmark model performance, latency, and cost

What you'll do

• Design and build production-grade GenAI applications using LangChain, LangGraph, and CrewAI
• Implement RAG pipelines with vector databases (Pinecone, Weaviate, ChromaDB) for enterprise knowledge systems
• Fine-tune and deploy LLMs (GPT-4, Claude, Llama) for domain-specific use cases
• Build multi-agent orchestration systems for complex workflow automation
• Develop evaluation frameworks to benchmark model performance, latency, and cost

What you'll need

• 2+ years working with LLMs, prompt engineering, or NLP in production environments
• Strong Python skills — comfortable with FastAPI, asyncio, and data pipelines
• Experience with at least one agent framework: LangChain, LangGraph, CrewAI, or AutoGen
• Understanding of embedding models, vector search, and retrieval-augmented generation
• Bonus: experience with PyTorch, model fine-tuning, or MLOps (MLflow, Weights & Biases)