Research Scientist - Generative AI & Foundation Models

About the Role

We are looking for an exceptional Research Scientist to lead research and development across NLP, Foundation Models, Generative AI, Reasoning Systems, Agentic AI, and Multimodal Intelligence.

The ideal candidate combines a strong research background with hands-on experience building, training, and deploying advanced AI systems. You will work at the intersection of fundamental research and product innovation to develop intelligent systems capable of reasoning, planning, retrieving knowledge, and operating autonomously across modalities.

Key Responsibilities

Research & Model Development

• Conduct original research in NLP, Generative AI, Foundation Models, Agentic AI, Reasoning, and Multimodal AI.
• Design novel model architectures, algorithms, and training methodologies.
• Train, fine-tune, evaluate, and optimize large language and foundation models.
• Work on instruction tuning, alignment, preference optimization, RLHF, RLAIF, DPO, GRPO, and reward modelling.
• Explore areas such as long-context models, RAG, multi-agent systems, memory, continual learning, synthetic data, AI safety, and knowledge grounding.

Productisation & Infrastructure

• Translate research breakthroughs into scalable, production-ready AI capabilities.
• Build end-to-end pipelines covering data curation, training, evaluation, deployment, monitoring, and continuous improvement.
• Collaborate with engineering and product teams to productionise research outcomes.
• Optimise distributed training, GPU utilisation, inference performance, and model serving.

Evaluation & Leadership

• Develop benchmarks for reasoning, hallucination reduction, retrieval quality, agent performance, safety, and user experience.
• Drive the technical strategy for advanced AI initiatives.
• Mentor researchers and ML engineers.
• Publish at leading conferences and represent the organisation across academic and industry forums.

Required Qualifications

• PhD or M.Tech/MS in Computer Science, AI, Machine Learning, NLP, Computational Linguistics, Data Science, or a related field.
• Strong publication record in conferences such as NeurIPS, ICML, ICLR, ACL, EMNLP, NAACL, COLM, or equivalent top-tier venues.
• Deep expertise in transformers, LLMs, representation learning, self-supervised learning, optimisation, PEFT, MoE, and long-context architectures.
• Experience with reasoning models, agentic workflows, planning, tool use, multi-agent systems, RAG, semantic search, vector databases, knowledge graphs, and memory systems.
• Strong understanding of multimodal and reinforcement-learning-based approaches.
• Expert proficiency in Python and PyTorch; experience with TensorFlow or JAX is preferred.
• Hands-on experience with multi-GPU training, distributed systems, Docker, Kubernetes, Linux, MLOps, and cloud AI infrastructure.

Preferred Qualifications

• First-author or highly cited publications, best-paper recognition, or prestigious fellowships such as PMRF, Google PhD Fellowship, Microsoft Research Fellowship, NVIDIA Fellowship, or JRF/SRF.
• Experience training or contributing to billion-parameter foundation models.
• Proven ability to translate research into production-grade AI products and measurable business impact.
• Strong open-source contributions, AI-related patents, technology transfer, or commercialisation experience.
• Demonstrated leadership in Agentic AI, Reasoning Systems, and the broader research community.

What Will Make You Stand Out

• Strong publication and citation record.
• Experience developing large-scale foundation models.
• Expertise in Agentic AI and advanced reasoning systems.
• Open-source or patent contributions.
• Ability to bridge scientific research, engineering, and product innovation.