Job: Audio Team Engineer (MLE 1/2)

About the job

Job Description: Audio Technology Engineer

Location: Kolkata/Bengaluru

About us

Founded in June 2016, Mihup is an AI-powered conversation intelligence platform that provides real-time personalized insights to contact center agents to improve sales and customer experience - resulting in revenue growth, reduced customer churn, and improved brand image. Our team is composed of engineers, machine learning scientists, and product specialists from leading institutions. Our clients include global Fortune 500 companies, and we are on track for aggressive expansion in India and globally. We are backed by Accel Partners and IdeaSpring Capital.

We are one of the few companies in India, who have trained and bulit their own entire conversational A.I. platform in-house. The conversation platform on the audio technology incorporates various deep learning based modules like diarization, speech recognition (ASR), noise reduction, speech enhancement, echo cancellation, text-to-speech, tonality etc.

About the Role

As a Team member of Audio Technology team, you will play a critical role in designing and building scalable, efficient and SOTA modules related to Audio and Speech processing. The developed modules is expected to process and analyze massive volumes of real-time and non-real-time interaction data (calls, real-time microphone feed etc,) to deliver actionable business insights and power real-time virtual agents or bots. You will collaborate with Team lead, Product Managers, Architects, and other Machine Learning Engineers to revamp architectures, identify new algorithms to build next generation of modules.

In this role, you’ll tackle key challenges such as achieving SOTA, building new cost-effective and yet highly accurate deep learning models, working with other engineers, and following best practices.

Key Problem Areas

- Achieving SOTA for various deep learning modules: Understanding problem areas, solutioning, defining data requirements, planning and deliveries.

- Deliverables: Understanding deliverables and accountability for quality of deliverables.

- Team Player: Acting as a team member, working with other engineers, to deliver required modules.

Key Responsibilities

- Develop and optimize speech recognition and audio processing algorithms

- Implement and test Audio Technology solutions

- Debug and optimize existing speech processing systems

- Write clean, maintainable, and efficient code

- Participate in code reviews and technical discussions

- Document code and technical specifications

- Collaborate with team members on integrated solutions

Expectations and skills:

- Bachelor's/Master's in Computer Science, Signal Processing, Electrical Engineering or related field

- Minimum 2 years of experience in speech/audio processing

- Strong programming skills in Python

- Experience with at least one deep learning framework (PyTorch/TensorFlow)

- Understanding of digital signal processing concepts

- Knowledge of speech processing fundamentals

Non-Functional Requirements: Strong focus on accuracy, scalability, and optimization in design and implementation.

Why This Role Matters

This role is pivotal in shaping the next-generation call analytics platform, addressing the critical need to scale and optimize systems to support 50x current volumes.