**Job Description:** **AI Lead Engineer** **Role Overview** We are seeking a **Lead** **Generative AI Engineer** with strong foundations in deep learning, transformer architecture, and practical experience building GenAI applications beyond basic RAG systems. The ideal candidate has hands-on experience/technical familiarity with LLM fine-tuning, multimodal models, retrieval systems, agentic frameworks, retrieval architectures, and production-grade ML deployment. This role will partner with engineering, data science, and CX teams to build intelligent agents, multimodal experiences, personalization systems, and knowledge-grounded AI solutions that power the future of customer engagement for global brands. ## **Key Responsibilities** ### **Generative AI, Multimodal Systems & Agentic Frameworks** - Build conversational and non-conversational, multimodal, and agentic AI applications using LLMs and frameworks such as LangChain, LangGraph, LlamaIndex, AutoGen, or similar. - Design AI workflows incorporating reasoning, planning, tool-use, memory, grounding, and external system integrations. - Develop Knowledge Graph (KG)-assisted AI systems, including entity extraction, linking, and KG-augmented retrieval. - Ensure safety, consistency, and hallucination-control through structured evaluation and guardrails. ### **Deployment, APIs & Cloud Engineering** - Transform models into scalable APIs and microservices using Python, FastAPI/Flask, Docker. - Deploy and monitor ML/AI systems in AWS/Azure/GCP, optimizing for cost, latency, and reliability. - Collaborate with MLOps teams on CI/CD pipelines, model versioning, monitoring, and automated evaluation. - Work with big data technologies including Apache Spark, Hadoop, and NoSQL databases such as MongoDB. ### **Model Development & Applied AI Engineering** - Build and optimize transformer-based and multimodal models using deep learning frameworks (e.g., PyTorch, TensorFlow). - Implement fine-tuning, alignment (RLHF/RLAIF), LoRA/QLoRA, pruning, and model evaluation pipelines. - Develop **information retrieval systems**, including hybrid dense–sparse retrieval, ranking, knowledge graphs, and relevance optimization. - Build predictive models and ML pipelines from scratch, including data preparation, feature engineering, and model selection. ### **Collaboration, Documentation & Mentorship** - Work cross-functionally with CX, engineering, and product stakeholders to translate business needs into AI solutions. - Document models, experiments, evaluation frameworks, and deployment processes. - Mentor junior engineers and contribute to internal best practices, reusable components, and R&D initiatives. ## **Required Technical Skills** - **Programming:** Python (advanced), SQL; robust experience with API development and data engineering, - **Backend Frameworks:** Flask, FASTAPI, Django - **Machine Learning:** Predictive modelling, deep learning, optimization, embeddings, vector search, model evaluation. - **Generative AI:** LLMs, RAG, multimodal architectures, agents, prompt engineering, grounding, knowledge graphs. - **Cloud Platforms:** AWS, Azure, or GCP with hands-on experience deploying and scaling AI systems. - **Data Technologies:** Apache Spark, Hadoop, MongoDB; strong understanding of data pipelines and large-scale processing. - **Math Foundations:** Linear algebra, probability, statistics. ## **Experience Requirements** - **Minimum 5-6 years** of hands-on software development experience including building and deploying machine learning models into production. - **2+ years of experience working with deep learning, GenAI**, or transformer-based architectures. - Demonstrated experience building GenAI applications **beyond simple RAG** (e.g., agents, multimodal, custom LLM fine-tuning). - Experience integrating AI systems in enterprise-grade environments. **Skill Category** **Lead AI Engineer** **Transformers & Deep Learning** Applies LoRA/QLoRA, distillation, debugging, optimization. **Generative AI (LLMs & Multimodal)** Builds tool-using pipelines, multilingual/multimodal flows. **Information Retrieval & Relevance** Implements hybrid retrieval + ranking, KG-enhanced semantic retrieval **Predictive Modeling** Builds and tunes end-to-end ML pipelines. **Knowledge Graphs** Builds KG pipelines (entity linking, embeddings). **Conversational AI** Multi-turn, multilingual dialogue systems with evaluation metrics. **Agentic Frameworks** Multi-step agent workflows with planning & memory. **Model Deployment** Scales services with CI/CD, monitoring, GPU/accelerator ops. **Cloud & MLOps** End-to-end model lifecycle automation. **Big Data & Pipelines** Uses Spark/Hadoop/MongoDB effectively. **Deep Learning** Understand and applied deep learning architectures – RNNs, LSTMs, Transformers ## **Attitude & Mindset** - Growth-oriented, collaborative, and experimentation-driven. - Strong problem-solving skills with a bias toward action. - Ability to communicate complex concepts clearly to non-technical stakeholders. - Open and flexible towards a hybrid work structure with no less than 2-days work from office – This is to ensure that the team working in the AI domain regularly connects and does knowledge exchange across projects **Location:** DGS India - Pune - Kharadi EON Free Zone **Brand:** Merkle **Time Type:** Full time **Contract Type:** Permanent