Develop and maintain low-latency streaming ASR systems for multiple Indian languages
Fine-tune and optimise speech models for medical terms, code-mixed speech and regional accents
Build machine translation systems that turn regional-language transcripts into Hindi or English
Build model-serving pipelines on GPU infrastructure and monitor latency and uptime
Develop clinical NLP for symptom extraction, specialist routing and summarisation
From the employer’s posting
Design, develop, and maintain low-latency streaming ASR systems supporting multiple Indian languages.
Fine-tune, evaluate, and optimise speech recognition models for clinical applications involving medical terminology, code-mixed conversations, and diverse regional accents.
Develop and maintain machine translation systems that convert regional-language speech transcripts into Hindi or English.
Build, deploy, and maintain model-serving pipelines on GPU-based infrastructure.
Collaborate with clinical teams to develop and improve models for symptom extraction, specialist routing, and clinical summarisation.
Min Experience: 3+ years Location: Bengaluru JobType: full-time
We are seeking a hands-on Senior Machine Learning (ML) & Automatic Speech Recognition (ASR) Engineer with strong expertise in production-grade machine learning, multilingual speech processing, and Medical Natural Language Processing (NLP). The role involves designing, developing, deploying, and optimising end-to-end speech AI pipelines for real-world clinical environments.
You will work on speech recognition and translation systems capable of handling noisy hospital outpatient departments (OPDs), low-bandwidth networks, diverse Indian languages, regional accents, and multilingual, code-mixed conversations. The ideal candidate will have practical experience deploying and scaling speech AI models, building GPU-based inference pipelines, and developing reliable solutions for healthcare applications.
Prior exposure to healthcare, healthtech, or health insurance is essential.
Requirements
Key Responsibilities
1. Multilingual ASR & Speech Recognition
Design, develop, and maintain low-latency streaming ASR systems supporting multiple Indian languages.
Fine-tune, evaluate, and optimise speech recognition models for clinical applications involving medical terminology, code-mixed conversations, and diverse regional accents.
Implement Voice Activity Detection (VAD), background noise suppression, and audio signal-processing techniques to improve transcription accuracy.
Optimise models for real-time inference through quantisation, streaming decoding, batching, and GPU memory management.
Develop offline and edge-deployable ASR pipelines for devices operating in low-connectivity environments.
Build evaluation frameworks and clinical audio datasets to continuously monitor recognition accuracy, inference latency, and overall model performance.
2. Machine Translation
Develop and maintain machine translation systems that convert regional-language speech transcripts into Hindi or English.
Fine-tune translation models using clinical text, medical terminology, symptom descriptions, dosage instructions, and doctors' notes.
Evaluate translation quality and ensure accurate handling of medical context and domain-specific terminology.
Identify and resolve inference bottlenecks to maintain production latency service-level agreements (SLAs).
3. Medical NLP & Clinical Intelligence
Collaborate with clinical teams to develop and improve models for symptom extraction, specialist routing, and clinical summarisation.
Enhance clinical NLP performance through medical entity recognition, medical coding, and structured information extraction from unstructured speech.
Integrate speech recognition outputs with downstream clinical NLP and intelligence pipelines.
Ensure data outputs align with HL7, FHIR, and Ayushman Bharat Digital Mission (ABDM) requirements.
Maintain traceability, auditability, and data integrity to support regulatory and insurance-related requirements.
4. Model Serving & GPU Infrastructure
Build, deploy, and maintain model-serving pipelines on GPU-based infrastructure.
Manage inference processes, worker configurations, service startup, logging, and operational reliability.
Maintain reproducible GPU environments, including framework versions, CUDA compatibility, and package dependencies.
Monitor inference latency, GPU utilisation, error rates, system uptime, and overall serving performance.
Evaluate on-premises versus cloud-based GPU infrastructure and contribute to infrastructure and inference cost optimisation.
Troubleshoot production issues and improve the scalability, stability, and efficiency of deployed models.
Must-Have Skills
Machine Learning
Automatic Speech Recognition (ASR)
Multilingual Speech Processing
Medical NLP
Speech Model Deployment and Optimisation
GPU-Based Model Serving
Required Experience
Strong hands-on experience developing and deploying production-grade machine learning and speech AI systems.
Practical experience with multilingual ASR, speech recognition, and translation pipelines.
Experience working with Indian languages, regional accents, and code-mixed speech is highly relevant.
Prior exposure to healthcare, healthtech, clinical technology, or health insurance is essential.
Experience optimising GPU-based inference systems and deploying models in real-world production environments.
Good-to-Have Skills
Voice Activity Detection (VAD) and audio signal processing.
Medical terminology, clinical entity recognition, and medical coding.
HL7, FHIR, and ABDM standards.
Edge AI and offline inference deployment.
GPU performance optimisation, quantisation, and streaming inference.
Clinical summarisation and structured data extraction.
Demand for ML Engineer roles
New ML Engineer postings at the employers we follow are down 4% over the last 6 weeks compared with the 6 before.
August 17Last week
Each bar is one week of new postings. The darkest bars are the last 6 weeks.
Counted from the hiring pages we’ve followed for at least five weeks, so employers we add don’t look like growth. Updated weekly.
About the role
We file this job under ML Engineer, by the work the posting describes.
ML engineers put machine learning models into products. They write the code that trains, ships, serves and monitors models so a recommendation or prediction works for real users.
Postings ask for strong Python and software engineering, a framework such as PyTorch, cloud services, and more and more often experience with large language models.
A data scientist often works out which model to build; an ML engineer makes it run reliably in production.
US government figures for computer and information research scientists
Median yearly pay, May 2025
$140,300
Expected change, 2025 to 2035
21.8% more jobsagainst 3.5% for all jobs
The government has no code for ML engineers, so these figures are for the closest group, computer and information research scientists.