Senior Agentic AI Engineer — Agentic Medical AI in Bengaluru, India

JazzX AI is seeking a talented Senior Agentic AI Engineer to join our team in Bengaluru, India. We are pioneering a new generation of clinical-grade AI agents, powered by a foundational model that seamlessly integrates 3D volumetric medical data with textual reasoning. These advanced agents will revolutionize various aspects of healthcare, including radiology workflows, treatment planning, case summarization, longitudinal patient analysis, and providing human-in-the-loop clinical decision support.

About the Role

In this pivotal role, you will be instrumental in designing the agent layer that transforms a powerful multimodal foundation model into a safe, reliable, and adaptive medical assistant. Your work will enable the analysis of 3D scans, retrieval of historical patient cases, explanation of complex medical findings, and the orchestration of sophisticated multi-step reasoning chains. This position is exceptionally suited for researchers with a strong background in multimodal learning, vision-language models (VLMs), or CLIP architectures who are eager to advance into autonomous medical reasoning, multimodal alignment, and contribute to real-world clinical impact.

What You Will Work On

• Architecting LLM-driven medical agents capable of:
• Reading and interpreting 3D medical volumes such as MRI, CT, and PET scans.
• Answering complex radiology questions, generating preliminary report drafts, and performing differential diagnosis reasoning.
• Retrieving similar prior cases using advanced contrastive embedding techniques.
• Synthesizing evidence from multiple instances across scans, laboratory results, and clinical notes.
• Executing intricate multi-step clinical workflows, such as image analysis leading to summary and recommendations.
• Integrating clinical agents with the 3D foundation model's representations to facilitate structured medical reasoning and maintain contextual memory across longitudinal patient data.
• Developing robust evaluation frameworks to ensure trustworthiness, interpretability, calibration, and clinical safety.
• Collaborating closely with clinicians and scientists to co-design tasks, curate datasets, and establish validation protocols.
• Benchmarking models against prominent open and proprietary datasets including MIMIC, CheXpert, MosMed, LiTS, BraTS, and India-specific clinical corpora.
• Publishing research findings at leading academic venues such as MICCAI, NeurIPS, AAAI, and EMNLP.
• Working with regulatory and product teams to translate cutting-edge research into medically viable and compliant systems.

Why This Role Appeals to CLIP / VLM / Chitrarth / Patram Authors

• This is a unique opportunity to work at the forefront of agentic LLMs, multimodal 3D vision, and clinical intelligence.
• Leverage your expertise in contrastive learning, cross-lingual modeling, multimodal alignment, and document understanding within the critical medical domain.
• Explore a greenfield space where medical agent design is still largely uncharted territory compared to 2D VLMs.
• Enjoy the freedom to publish your work, contribute to open-source initiatives, and propose novel clinical reasoning benchmarks.
• Make a tangible impact beyond academia by directly improving physician workflows and patient outcomes.
• Contribute to the creation of India's first medical foundation-model agent system, setting new global standards in healthcare AI.

What We're Looking For

• Proven experience in developing LLM-based agents, tool-using models, chain-of-thought systems, or orchestrated AI pipelines.
• A strong background in multimodal learning, vision–language transformers, contrastive models, or clinical Natural Language Processing (NLP).
• Familiarity with medical imaging modalities, radiology workflows, DICOM pipelines, or 3D data representations.
• Expertise in PyTorch/JAX and experience with large-scale model training and inference.
• The ability to design safe, interpretable, and auditable AI reasoning systems.
• A strong publication record in areas such as multimodal AI, medical AI, agentic systems, or Large Language Models (LLMs).
• Comfort and effectiveness in collaborating with clinicians and biomedical researchers.

Nice to Have

• Experience with popular agent frameworks like LangChain, LLaMA Index, Haystack, or custom pipeline development.
• A background in clinical ontologies, SNOMED, RadLex, or UMLS.
• Prior contributions to significant medical AI challenges (e.g., BraTS, RSNA, MedVQA, MedIC).
• Experience working with multilingual or code-mixed medical text.
• Demonstrated ability to transition research concepts into production-grade systems.

What We Offer

• Competitive compensation package.
• Access to high-compute clusters and extensive proprietary datasets.
• Full autonomy to publish research and define development roadmaps.
• Dynamic cross-functional collaboration with clinicians, regulatory experts, and product teams.
• A clear and impactful path to building systems that meaningfully enhance healthcare outcomes.