About Nuvo AI

Nuvo AI Pvt. Ltd., part of the Meril Group, is focused on developing AI-driven technology solutions and intelligent software products. Our teams work on innovative applications that combine artificial intelligence, data engineering, and enterprise software to solve complex business challenges.

Key Responsibilities

1. Generative AI and LLM Engineering

- Design, develop, and deploy LLM-powered applications using commercial and open-source models such as GPT, Claude, Gemini, and other suitable foundation models.
- Build advanced AI solutions for patent analysis, document understanding, summarization, technical question answering, and automated research.
- Develop prompt engineering, structured generation, tool calling, and context-management strategies.
- Evaluate, adapt, and fine-tune language models to improve performance on domain-specific tasks.

2. RAG, Semantic Search and Information Retrieval

- Design and optimize end-to-end RAG pipelines for large-scale patent and technical document collections.
- Develop document ingestion, parsing, chunking, embedding generation, vector search, hybrid retrieval, reranking, and answer-generation workflows.
- Improve search relevance and contextual accuracy for complex patent and technical queries.
- Build intelligent retrieval systems for prior-art discovery, patent similarity, claim analysis, and technology landscape research.
- Implement retrieval evaluation and grounded-answer validation to minimize irrelevant results and LLM hallucinations.

3. Agentic AI and MCP Integration

- Build AI agents and multi-agent workflows for autonomous research, information retrieval, analysis, and task execution.
- Develop agent orchestration using frameworks such as LangGraph, LangChain, or equivalent technologies.
- Implement Model Context Protocol (MCP) servers and tool integrations to connect AI systems with search engines, databases, APIs, and enterprise services.
- Design reliable tool-use workflows with appropriate validation, error handling, access controls, and human approval mechanisms.
- Optimize agent planning, context management, task completion, and execution reliability.

4. NLP and Patent Document Intelligence

- Develop NLP and deep learning solutions for patent classification, entity extraction, technical concept identification, and document similarity.
- Build intelligent workflows for analyzing patent claims, abstracts, descriptions, and technical literature.
- Develop AI-driven patent categorization, novelty search, landscape analysis, and automated reporting capabilities.
- Process and analyze large volumes of structured and unstructured documents.
- Apply transformer models, embeddings, and other relevant techniques to improve document understanding and information discovery.

5. Model Training, Evaluation and Optimization

- Fine-tune or post-train foundation models using suitable techniques such as supervised fine-tuning, LoRA, or preference optimization where appropriate.
- Develop systematic evaluation frameworks for retrieval quality, answer faithfulness, classification performance, and agent reliability.
- Optimize inference latency, throughput, token consumption, GPU utilization,



and serving costs.
- Evaluate model performance through offline benchmarks, production monitoring, and feedback-driven improvements.
- Apply caching, batching, quantization, context compression, and model routing where appropriate.

6. AI Deployment and Production Engineering

- Develop scalable AI services and REST APIs using Python and FastAPI or equivalent frameworks.
- Deploy AI applications on cloud platforms such as AWS or GCP, using services including AWS Bedrock and Google Cloud Vertex AI where appropriate.
- Containerize and deploy services using Docker and relevant orchestration technologies.
- Integrate AI pipelines with vector databases, Elasticsearch/OpenSearch, MongoDB, PostgreSQL, and other data systems.
- Implement monitoring, logging, version control, automated testing, CI/CD, and model lifecycle management.
- Ensure AI systems are reliable, secure, maintainable, and suitable for enterprise use.

7. Technical Ownership and Collaboration

- Take ownership of AI features from requirement analysis and prototyping through implementation, deployment, and continuous improvement.
- Collaborate with backend engineers, product teams, and other technical stakeholders to integrate AI capabilities into the platform.
- Review code, evaluate technical approaches, and contribute to AI architecture and engineering best practices.
- Research emerging AI technologies and assess their practical value for the Project.
- Mentor team members and provide technical guidance based on experience and demonstrated expertise.

Required Experience

- 2–6+ years of relevant experience in AI/ML Engineering, Data Science, Applied AI, NLP, or Generative AI, depending on the depth of hands-on expertise.
- Strong practical experience developing AI/ML applications using Python.
- Hands-on experience with LLMs, RAG pipelines, embeddings, semantic search, or intelligent information retrieval.
- Experience building AI applications using frameworks such as LangChain, LangGraph, LlamaIndex, or equivalent tools.
- Experience developing APIs and integrating AI systems with databases, search engines, or enterprise applications.
- Understanding of model evaluation, prompt optimization, retrieval quality, and production deployment.
- Demonstrated ability to independently solve technical problems and deliver working AI solutions.

Candidates with deeper experience in production architecture, advanced agentic systems, model fine-tuning, GPU inference, or technical leadership will be considered for more senior responsibilities.

Educational Qualifications

- Bachelor's or Master's degree in Computer Science, Information Technology, Artificial Intelligence, Machine Learning, Data Science, Electronics and Communication Engineering, or a related technical discipline.
- A postgraduate qualification in AI/ML,



Data Science, Computer Science, or a related field is preferred but not mandatory.
- Strong fundamentals in machine learning, deep learning, algorithms, statistics, and data structures.

Mandatory Technical Skills

Programming and AI/ML

- Python and SQL.
- PyTorch, TensorFlow, or Scikit-learn.
- Machine learning, deep learning, NLP, transformers, and model evaluation.

Generative AI and LLMs

- Practical experience with LLMs, prompt engineering, embeddings, and structured outputs.
- RAG pipeline development and optimization.
- Familiarity with commercial LLM APIs and/or open-source foundation models.

Search and Retrieval

- Vector embeddings, vector databases, and semantic search.
- Document processing, chunking, indexing, and retrieval optimization.
- Understanding of similarity search, hybrid retrieval, and ranking.

AI Frameworks and Integration

- LangChain, LangGraph, LlamaIndex, or equivalent frameworks.
- FastAPI, REST APIs, and database integration.
- Git and collaborative software development practices.

Deployment and Engineering

- Docker and familiarity with AWS, GCP, or Azure.
- Model deployment, logging, monitoring, testing, and performance evaluation.
- Understanding of secure and scalable AI application design.

Preferred / Good-to-Have Skills

- Patent Intelligence and Legal-Tech: Experience with patent search, prior-art analysis, claim analysis, patent classification, IP analytics, or technical document intelligence.
- Advanced Agentic AI: Multi-agent systems, agent planning, tool use, memory, workflow orchestration, and autonomous task execution.
- Model Context Protocol (MCP): Building MCP servers and integrating AI agents with search tools, databases, and enterprise applications.
- Advanced RAG: Hybrid retrieval, reranking, GraphRAG, knowledge graphs, query expansion, and retrieval evaluation.
- LLM Fine-Tuning: Supervised fine-tuning, LoRA/QLoRA, PEFT, preference optimization, and domain adaptation.
- Inference Optimization: vLLM, NVIDIA Triton, TensorRT-LLM, ONNX Runtime, GPU acceleration, quantization, and distributed inference.
- Cloud AI Platforms: AWS Bedrock, Lambda, S3, EC2, Google Cloud Vertex AI, or equivalent managed AI services.
- Search and Databases: Elasticsearch, OpenSearch, FAISS, Qdrant, Milvus, MongoDB, PostgreSQL, and Redis.
- MLOps and Infrastructure: MLflow, CI/CD, Kubernetes, Ray, experiment tracking, model monitoring, and production observability.
- AI Safety and Governance: Prompt-injection mitigation, data privacy, secure tool execution, access controls, and human-in-the-loop validation.

Why Join Nuvo AI?

- Build advanced AI capabilities for an AI-powered Patent Intelligence Platform.
- Work on practical applications of Generative AI, LLMs, RAG, NLP, and agentic AI.
- Solve challenging problems involving large-scale technical documents and intelligent information retrieval.
- Contribute to AI architecture, model evaluation, and production deployment.
- Collaborate with engineering and product teams to deliver creative AI-driven solutions.
- Help shape the next generation of AI-powered patent research and intelligence tools.

📌 Senior AI Engineer (Vapi)
🏢 Nuvo AI
📍 Vapi