Job Summary:We are seeking a highly technical and results-driven Machine Learning Engineer to design and build cutting-edge solutions for voice and text-based systems. You'll work on complex, unsolved problems using advanced ML techniques, with a strong focus on speech-native models, multimodal learning, and hardware optimization. If you're passionate about conversational AI, LLM-based virtual assistants, and real-world implementation, this role is for you.
Job Responsibilities:- Design and implement an end-to-end LLM-based conversational speech virtual assistant
- Benchmark and evaluate speech-native models (e.g., Moshi, SesameAI) for in-vehicle applications
- Fine-tune models for automotive domain adaptation
- Develop API frameworks for vehicle system control
- Optimize for Qualcomm SA8255P hardware platform
- Maintain and write Python code for audio preprocessing and integration
- Document model architectures, benchmarks, and optimization strategies
- Execute the full modeling lifecycle: data cleansing, feature engineering, and model selection
- Collaborate in an Agile Scrum team to prototype, build, and deploy real-world ML solutions
Required Skills:- 3+ years of hands-on ML experience in industry
- Deep knowledge of Voice2Voice architectures and speech-native models
- Experience with model quantization and edge-device optimization
- Strong audio processing expertise (feature extraction, acoustic modeling, noise handling)
- Proficiency with TensorFlow or PyTorch
- Solid understanding of ASR and TTS technologies, including transformer variants
- Experience with multimodal learning, attention mechanisms, and cross-modal fusion
- Familiarity with tools like TorchAudio and Librosa
- Hands-on experience fine-tuning transformer models using Huggingface, PyTorch/TensorFlow
- Experience with distributed training pipelines
- Background in algorithm design and complexity analysis
- Strong problem-solving, decision-making, and analytical skills
- Demonstrated ability to learn new technologies and collaborate across teams
- Proven track record of delivering high-impact, measurable outcomes
Preferred Skills:- Research experience in transformers or multimodal systems (academic or industry)
- Experience developing systems for in-vehicle applications
Education:Not specified - assumed Bachelor's or master's in computer science, AI, or related field
