Machine Learning Engineer – Medical AI Agents (RAG, SAG, Generative & Agentic Frameworks)

Hyderabad

Founded by highly respected Silicon Valley veterans - with its design centers established in Santa Clara, California. / Hyderabad/ Bangalore

Machine Learning Engineer – Medical AI Agents (RAG, SAG, Generative & Agentic Frameworks)

Location:

Hyderabad, India (Hybrid/Remote options available)

About the Role

We are building next-generation AI Agents for Clinical Decision Support—systems that combine the power of LLMs, RAG/SAG architectures, and agentic reasoning to assist doctors, reduce burnout, and extend quality care to millions. As a Machine Learning Engineer, you will work at the frontier of retrieval-augmented, structured-augmented, and agentic AI systems, designing medical copilots that can converse, reason, and act safely in real-world healthcare settings.

You’ll collaborate with clinicians, public health bodies, and AI researchers to bring these agents from lab to hospital floor, tailored for multilingual, culturally diverse, and resource-constrained environments.

Key Responsibilities

• Develop and deploy intelligent Medical AI Agents using RAG (Retrieval-Augmented Generation) and SAG (Structured-Augmented Generation) to assist in diagnostics, triage, therapy planning, and patient communication

• Architect multi-agent LLM workflows that reflect real clinical roles (e.g., GP, nurse, pharmacist), with memory, task decomposition, and role-specific context

• Fine-tune and align LLMs with clinical reasoning datasets using SFT, RLHF, and preference modeling to ensure high factuality and low hallucination rates

• Design multimodal integration pipelines that ingest EHRs, lab reports, imaging summaries, and patient-reported data for robust contextual reasoning

• Collaborate directly with doctors, nurses, and medical administrators to validate agent outputs and refine interaction models

• Participate in user studies, clinical pilots, and co-design workshops to ensure usability and trustworthiness of deployed agents

• Engage with public and private healthcare partners across India, Southeast Asia, and multilingual regions to customize deployment and localization

• Build frameworks for auditability, explainability, and fallback behavior in line with SaMD (Software as a Medical Device) and CDSCO/FDA guidance

• Incorporate empathic communication techniques, including tone modulation, uncertainty handling, and culturally sensitive phrasing, into agent behavior

• Benchmark systems on MedQA, PubMedQA, MedMCQA, and internal evaluations to continuously improve real-world performance

Key Qualifications

Educational Background:

• Master’s or higher in Computer Science, AI/ML, Biomedical Informatics, or related field from a premier institution (e.g., IISc, IITs, IIIT-H, BITS Pilani, or top global universities)

• Outstanding Bachelor’s candidates considered with strong track record in LLM research, clinical AI projects, or medical NLP deployment

• Bonus: Coursework or thesis in Human-Centered AI, Medical Ethics, Cognitive Science, or Clinical Informatics

Technical Expertise:

• Strong command of transformer models, RAG/SAG architectures, contextual embeddings, and retrieval orchestration tools (e.g., LangChain, LlamaIndex)

• Proficiency in Python, PyTorch, HuggingFace, vector DBs (FAISS, Qdrant), and EHR integration frameworks (FHIR, HL7)

• Experience designing agentic LLM frameworks with memory persistence, intent handling, and inter-agent communication

• Exposure to multimodal reasoning, including structured medical data (labs, symptoms) and unstructured text (progress notes, prescriptions)

• Familiarity with medical ontologies (SNOMED CT, UMLS, ICD-10) and working with large-scale clinical text corpora

• Understanding of clinical safety, bias risks, and hallucination controls for generative systems

• Experience with RLHF, zero/few-shot prompt tuning, and retrieval grounding in high-stakes settings

Collaboration & Stakeholder Engagement

• Proven ability to collaborate with healthcare professionals, understand their mental models, and iteratively translate them into agent logic

• Participated in field testing, simulation environments, or real-world deployments in clinical, telemedicine, or public health settings

• Strong communication skills to present and defend model behavior to both technical and non-technical stakeholders

• Willingness to engage in design sprints, qualitative research, and in-clinic observations

• Empathy for users under pressure—your systems will support people making life-saving decisions

Preferred Attributes

• Experience building for multilingual contexts or deploying in LMIC (Low- and Middle-Income Country) healthcare systems

• Strong understanding of AI safety, uncertainty quantification, and fallback design

• Familiarity with regulatory compliance in medical AI (e.g., SaMD, CDSCO, HIPAA, GDPR)

• Published research or open-source contributions in medical NLP, generative agents, or multi-agent LLM frameworks

Experience

• Up to 15 years

• 6–8 years of experience in AI/ML (with a strong focus on NLP/LLMs/RAG frameworks), including:

• At least 2–3 years in independently delivering production-grade AI systems

• At least 1–2 years in a tech-lead or mentoring capacity, ideally in a startup, research lab, or interdisciplinary team

• Hands-on experience with RAG/SAG pipelines, LLM fine-tuning, and agent orchestration, not just model usage

Contact:

Uday

Mulya Technologies

muday_bhaskar@yahoo.com

"Mining The Knowledge Community"