At pmspace.ai, we're building next-generation AI tools for project management intelligence. Our platform leverages graph databases, NLP, and large language models (LLMs) to transform complex project data into actionable insights. Join us to pioneer cutting-edge solutions in a fast-paced, collaborative environment.
Role Overview
We seek a Python Developer with expertise in graph databases (Neo4j), RAG pipelines, and vLLM optimization. You'll design scalable AI systems, enhance retrieval-augmented workflows, and deploy high-performance language models to power our project analytics engine.
Key Responsibilities
Architect and optimize graph database systems (Neo4j) to model project knowledge networks and relationships.
Build end-to-end RAG (Retrieval-Augmented Generation) pipelines for context-aware AI responses.
Implement and fine-tune vLLM for efficient inference of large language models (LLMs).
Develop Python-based microservices for data ingestion, processing, and API integrations (FastAPI, Flask).
Collaborate with ML engineers to deploy transformer models (e.g., BERT, GPT variants) and vector databases.
Monitor system performance, conduct A/B tests, and ensure low-latency responses in production.
Required Skills
Proficiency in Python and AI/ML libraries (PyTorch, TensorFlow, Hugging Face Transformers).
Hands-on experience with graph databases, especially Neo4j (Cypher queries, graph algorithms).
Demonstrated work on RAG pipelines (retrieval, reranking, generation) using frameworks like LangChain or LlamaIndex.
Experience with vLLM or similar LLM optimization tools (quantization, distributed inference).
Knowledge of vector databases (e.g., FAISS, Pinecone) and embedding techniques.
Familiarity with cloud platforms (AWS/GCP/Azure) and containerization (Docker, Kubernetes).
Job Type: Full-time
Pay: ?5,000.00 - ?7,000.00 per month
Schedule:
Day shift
Work Location: Remote
Expected Start Date: 01/08/2025
Beware of fraud agents! do not pay money to get a job
MNCJobsIndia.com will not be responsible for any payment made to a third-party. All Terms of Use are applicable.