JD: AI/ML Python Engineer RAG & LLM Systems
Location: Pune (On-Site)
Experience: 4+ years
Job Description
We are looking for a highly skilled AI/ML Python Engineer with hands-on experience in LLMs (Llama), RAG pipelines, vector databases, and backend API development. The ideal candidate is strong in Python, machine learning workflows, embeddings, and scalable backend architecture, with the ability to build reliable AI-driven applications end-to-end.
Key Responsibilities
AI/ML Development & Optimization
- Build and integrate GenAI and classical ML models using LangChain, LlamaIndex, Scikit-learn, TensorFlow, PyTorch.
- Optimize models via hyperparameter tuning, cross-validation, and monitor performance using observability tools.
- Improve accuracy, latency, and reliability of LLM-based pipelines.
RAG, Embeddings & Vector Search
- Implement RAG workflows, prompt-engineered LLM interactions, and embeddings-based similarity search.
- Work with vector DBs such as Pinecone, Weaviate, Supabase, PGVector.
Multimodal Data Processing
- Preprocess and manage text, image, and video datasets.
- Build lightweight multimodal pipelines for feature extraction and inference.
Backend & API Engineering
- Build scalable backend services using FastAPI, Django, Flask.
- Design clean RESTful APIs and integrate AI APIs, external SDKs, and authentication workflows.
Database & Data Workflow Management
- Work with PostgreSQL, MySQL, MongoDB and integrate vector search layers.
- Support ETL workflows and backend data pipelines.
Task Automation & Scheduling
- Implement background tasks using Celery, Redis queues, and cron-based scheduled jobs.
Containerization & Deployment
- Package and deploy applications using Docker across dev, staging, and production environments.
End-to-End Development
- Contribute across the SDLCfrom requirement analysis, design, development, testing, deployment, and optimization.
Required Skills
- Strong Python expertise
- Hands-on with Llama, LangChain, LlamaIndex
- Experience with Vector Databases (Pinecone, Weaviate, PGVector, etc.)
- Machine Learning & Deep Learning (Sklearn, TF, PyTorch)
- REST API development (FastAPI/Django/Flask)
- Docker, Celery, SQL/NoSQL Databases
- Knowledge of RAG, embeddings, prompt engineering
- Experience working with multimodal datasets
Nice to Have
- Experience with MLOps tools (MLflow, Weights & Biases)
- Cloud: AWS / GCP / Azure
- CI/CD pipelines
- Experience fine-tuning LLMs or building custom embeddings models
