Senior AI Engineer

Teleradiology Solutions Pvt. Ltd.; Bengaluru; On-site; INR 2,400,000 - 4,200,000; Full time

Job summary

Daignostix, the AI arm of Teleradiology Solutions, Bengaluru, builds clinical AI for medical imaging and deploys it in hospital settings. You will own AI algorithms end-to-end, from problem framing and data handling to modelling, clinical evaluation, and production deployment with radiologist collaboration.

The role focuses on hands-on engineering for production-grade AI systems, including GPU inference, monitoring, and on-prem/cloud deployment, without people-management duties.

Qualifications

• B.Tech/B.E./M.Tech/MS/M.Sc. from IIT, IISc, IIIT, BITS Pilani or comparably selective institution, in CS, ECE, AI, Data Science, Math, Stats or related quantitative field.
• 5-8 years of hands-on AI/ML engineering, including production deployment of deep-learning systems.
• Substantive AI/ML engineering experience at leading medical-imaging AI company or top AI product/research org with production ownership.
• Able to walk through at least two AI/ML systems you drove from data to deployment with measurable outcomes.
• Expert Python, PyTorch and/or TensorFlow/Keras, and working proficiency in NumPy, scikit-learn and OpenCV.
• Strong command of deep learning, CV, classical ML, statistics, calibration and experimental design.
• Linux, Git, Docker, FastAPI/REST, pytest, CI/CD, logging/monitoring, model/data versioning, AWS/Azure/GCP.

Responsibilities

• Own one or more clinical AI algorithms from proposal through data strategy, modelling, validation, deployment, monitoring and improvement.
• Design and train CNN, transformer and hybrid models for classification, detection and segmentation.
• Architect DICOM pipelines and dataset curation with lineage and versioning; address leakage and bias.
• Define evaluation protocols, metrics, calibration and subgroup analyses with radiologists.
• Turn validated models into production services (Docker, FastAPI/REST) with robust GPU inference and monitoring.
• Deploy on AWS, Azure or GCP and on on-prem GPU infra; track experiments and models; drift monitoring.
• Lead design/code/model-readiness reviews; mentor junior engineers; set engineering standards; no people management.

Skills

Python

PyTorch

TensorFlow/Keras

NumPy

scikit-learn

OpenCV

Deep Learning

CNNs

Vision Transformers

DICOM / NIfTI

Education

B.Tech/B.E./M.Tech/MS/M.Sc. from IIT/IISc/IIIT/BITS Pilani or comparable institute

Tools

Docker

FastAPI

REST

pytest

CI/CD

AWS

Azure

GCP

Git

Monitoring

Model versioning

ONNX/TensorRT

Job description

Bangalore North, India | Posted on 10/26/2026

About Teleradiology Solutions / Daignostix

Teleradiology Solutions (TRS), founded in 2002 by two Yale-trained physicians and headquartered in Ardmore, PA, provides 24x7 diagnostic radiology reporting to over 150 hospitals across 21 countries, through a network of radiologists (US, Indian certified) and technology-driven workflows.

Daignostix, the AI arm of TRS based in Whitefield, Bengaluru, builds clinical AI for medical imaging and healthcare workflows. TRS also maintains an active academic and research focus, including AI in radiology ( www.radguru.net ).

Role Summary

Daignostix builds clinical AI for medical imaging, with algorithms spanning neuro and stroke, pulmonary embolism, aortic pathology, urology, spine and rib fracture, solid-organ trauma and paediatric chest. Our models run on real patient studies alongside radiologists (US, Indian certified), so success means clinical accuracy, robustness, latency and safe operation in production.

As a Senior AI Engineer, you will own clinical AI algorithms end to end: problem formulation, data and ground truth, modelling, clinical evaluation, production deployment and monitoring. This is a hands-on senior individual-contributor role for an engineer who has already shipped and operated production AI systems.

Roles and Responsibilities

• End-to-end algorithm ownership. Own one or more clinical AI algorithms from proposal through data strategy, modelling, validation, deployment, monitoring and improvement. Define the technical plan, flag risks early and drive decisions to closure.
• Model development. Design and train CNN, transformer and hybrid models for classification, detection and segmentation, using classical ML where it is the better fit.
• Data and ground truth. Architect DICOM pipelines and dataset curation with lineage and versioning. Catch leakage, label-quality issues, cohort bias and scanner/site variability before they become model problems.
• Clinical evaluation. Define evaluation protocols, clinical metrics, calibration and subgroup analyses. Lead false-positive/false-negative review and threshold selection with radiologists (US, Indian certified).
• Production engineering. Turn validated models into production services (Docker, FastAPI/REST) with robust GPU inference, observability and safe failure behaviour. Optimise with ONNX, TensorRT or CUDA where it measurably helps without changing clinical outputs.
• MLOps and reliability. Deploy on AWS, Azure or GCP and on on-premise GPU infrastructure. Own experiment tracking, model registries, CI/CD, regression and golden-set testing, drift monitoring and rollback.
• Technical leadership. Lead design, code and model-readiness reviews, mentor junior engineers and set engineering standards. This role has no people-management responsibility.

Requirements

Must-have

• Education. B.Tech/B.E./M.Tech/MS/M.Sc. from an IIT, IISc, IIIT, BITS Pilani or a comparably selective institution, in Computer Science, Electrical/Electronics Engineering, AI, Data Science, Mathematics, Statistics or a related quantitative field, with a strong academic record.
• Experience. 5-8 years of hands-on AI/ML engineering, including multiple years taking deep-learning systems into production. Medical-imaging or healthcare AI experience is strongly preferred.
• Product pedigree. Substantive AI/ML engineering experience at a leading medical-imaging AI company or a top-tier AI product or research organisation, with clear ownership of production AI products.
• Proof of production ownership. Able to walk through at least two AI/ML systems you personally drove from data through deployment, monitoring and iteration, with measurable outcomes.
• Programming. Expert Python, deep PyTorch and/or TensorFlow/Keras, and working proficiency in NumPy, scikit-learn and OpenCV.
• ML depth. Strong command of deep learning, computer vision, classical ML, statistics, calibration and experimental design.
• Production engineering. Linux, Git, Docker, FastAPI/REST, pytest, CI/CD, logging and monitoring, model/data versioning, and hands-on AWS, Azure or GCP.

Good-to-have

• Scale and performance. Kubernetes, distributed training/inference, ONNX/TensorRT, CUDA profiling and infrastructure-as-code.
• Medical imaging tooling. DICOM/NIfTI, MONAI and nnU-Net.
• Multimodal AI. VLMs, LLMs, RAG and PEFT, as an addition to, not a substitute for, core deep-learning depth.

Key Skills

Python | PyTorch | TensorFlow / Keras | NumPy | scikit-learn | OpenCV | Classical ML | CNNs | Vision Transformers | Detection | Segmentation | DICOM / NIfTI | MONAI / nnU-Net | Clinical Metrics | Calibration | Docker | FastAPI | pytest | CI/CD | MLflow / Weights & Biases | AWS / Azure / GCP | Kubernetes | ONNX / TensorRT

Why Join Daignostix

• Clinical AI that matters. Algorithms evaluated on real patient studies, where robustness, latency and reliability have direct operational consequences.
• Direct access to radiologists. Work with subspecialist radiologists (US, Indian certified) on annotation strategy, threshold decisions and clinical failure analysis.
• Senior influence, hands-on scope. Shape architecture and technical standards across GPU infrastructure, cloud and on-premise deployment, while staying close to code and data.