About the job AI/ML Engineer - Generative AI Music
About Mindtech
Mindtech is your gateway to exciting and impactful tech projects. We specialize in end-to-end software outsourcing, linking Latin American talent with global opportunities. Our fast, cost-effective approach ensures that our clients receive exceptional service and innovative solutions. With a diverse team of over 70 skilled professionals across Latin America and the US, we are committed to delivering software that drives success.
Title: AI/ML Engineer - Generative AI Music
Location: Remote / Flexible
Job Type: part-Time / Contract
About the Project:
We are developing an innovative Generative AI Music project, pushing the boundaries of AI-driven music creation and enhancement. Our goal is to deeply understand and optimize generative models, such as Riffusion, to achieve the highest possible audio quality.
Key Responsibilities:
Conduct in-depth exploratory data analysis (EDA) to uncover patterns, biases, and improvement opportunities in generative AI models.
Develop and refine hypothesis-driven experiments to assess and improve audio quality and generation strategies.
Analyze audio data using statistical and signal processing techniques to evaluate model performance.
Work closely with engineers to provide insights that inform architectural improvements in AI-generated music models.
Apply statistical methods and data-driven techniques to measure and enhance model quality.
Collaborate with cross-functional teams to translate findings into practical enhancements for generative audio models.
Implement, optimize, and maintain generative audio models running on AWS SageMaker, including monitoring performance and deploying model updates.
Required Qualifications:
3+ years of experience in AI/ML engineering or data science with strong involvement in deploying and maintaining ML models in production.
Solid background in audio and/or image data analysis, with demonstrated use of statistical techniques to improve model performance.
Proficiency in Python and ML frameworks (NumPy, SciPy, Pandas, PyTorch, TensorFlow).
Experience with AWS SageMaker (training, tuning, and deploying models).
Strong understanding of signal processing and time-series techniques relevant to audio data.
Ability to experiment with data-driven approaches and translate findings into scalable, production-ready solutions.
Preferred Qualifications (Nice to Have):
Familiarity with spectrogram analysis (e.g., Mel-spectrograms, STFT) is a plus but not requiredcandidates can learn on the job.
Some experience with neural vocoders (e.g., WaveNet, MelGAN, HiFi-GAN) or a willingness to learn.
Experience with audio post-processing techniques, such as noise reduction, equalization, or dynamic range processing.
Knowledge of data augmentation techniques for improving model robustness.
Experience working with generative AI models, particularly in audio or music applications.
Understanding of multimodal AI approaches combining image, audio, and text data.
We offer
100% remote work
Salary in USD
Referral Program
Location: You can apply if are based at Latam´s countries