Senior Gen AI Engineeer

We are looking for an experienced Senior Generative AI Engineer to join our Core AI team. You will lead the design, development, and deployment of enterprise-grade Generative AI solutions ranging from advanced Retrieval-Augmented Generation (RAG) pipelines to autonomous multi-agent systems and fine-tuned LLMs. You will bridge the gap between cutting-edge AI research and scalable cloud software, working closely with Product, Data, and DevOps teams to deliver high-impact, low-latency AI features.

The candidate will have responsibilities across the following functions:

Architecting and Building GenAI Systems:

• Design, build, and maintain production-grade GenAI applications (RAG pipelines, semantic search, AI agents, dynamic prompt flows).
• Build multi-agent workflows using frameworks like LangChain, LangGraph, LlamaIndex, AutoGen, or CrewAI.
• Implement robust evaluation metrics (e. g., Ragas, TruLens) for hallucination detection, response quality, and safety guardrails (Guardrails AI, NeMo).

Model Fine-Tuning and Optimisation:

• Fine-tune open-source Foundation Models (e. g., Llama, Mistral, Qwen) using techniques like LoRA, QLoRA, and PEFT.
• Optimise model latency and serving costs using tools like vLLM, TensorRT-LLM, Ollama, or Triton Inference Server.
• Implement embedding models, dynamic chunking strategies, and vector space optimisations.

Data and Vector Infrastructure:

• Manage vector databases at scale (Pinecone, Qdrant, Milvus, Weaviate, or Pgvector).
• Implement structured hybrid retrieval strategies combining dense vector search with sparse kEYword search (BM25) and re-ranking models (Cohere, BGE).

Engineering and Deployment:

• Write clean, maintainable, and asynchronous Python microservices (FastAPI / Flask).
• Package and deploy AI microservices using Docker, Kubernetes, and cloud AI platforms (AWS Bedrock/SageMaker, Azure OpenAI, GCP Vertex AI).
• Set up LLMOps practices, including observability, tracing, and token monitoring (LangSmith, Phoenix, Weights & Biases).

Requirements:

• Experience: 5+ years of software engineering / Machine Learning experience, with 2+ years dedicated specifically to Generative AI & LLMs.
• Core Language: Strong command of Python (async programming, pydantic, FastAPI).
• LLM Frameworks: Hands-on expertise with LangChain, LangGraph, LlamaIndex, or AutoGen.
• Vector DBs & RAG: Deep experience with Vector Databases, hybrid search, graph RAG, and re-ranking techniques.
• Fine-Tuning: Practical experience with Hugging Face ecosystem, PEFT, LoRA/QLoRA, and dataset curation.
• Serving & Inference: Experience with high-throughput serving engines like vLLM or TensorRT-LLM.
• Cloud & DevOps: Proficiency in AWS, Azure, or GCP, along with Docker, CI/CD, and basic MLOps/LLMOps tools.
• Data Structures: Solid understanding of software architecture, system design, and API design principles.

Nice-to-Have Skills:

• Experience with Multimodal AI models (Vision-Language models, audio processing).
• Background in classical ML / NLP (spaCy, NLTK, transformers) before the LLM era.
• Contributions to open-source GenAI frameworks or projects.
• Exposure to local AI deployments (e. g., ONNX, Ollama, edge execution).
• Degree: Bachelor's or Master's in Computer Science, Data Science, Artificial Intelligence, or a related quantitative field (or equivalent practical experience).