Position Overview
We are seeking a Senior Machine Learning Engineer to design, deploy, and optimize our next-generation Conversational AI and Data Analytics platforms. You will bridge the gap between AI research and production engineering. Your primary focus will be optimizing and scaling core Speech (ASR, TTS) and Language Model (LLM, SLM) pipelines across hybrid cloud and local edge environments.
Key Responsibilities
Pipeline Deployment & Architecture
- Deploy AI Pipelines: Build production-grade, low-latency pipelines for ASR, TTS, and Small Language Models (SLMs).
- Hybrid Deployment: Manage deployment topologies across multi-cloud environments and bare-metal local hardware.
- API Development: Create high-performance, asynchronous REST and WebSocket APIs using FastAPI to serve real-time conversational agents.
MLOps & Infrastructure
- CI/CD Automation: Design automated machine learning pipelines for model testing, versioning, and continuous deployment.
- Containerization: Pack applications using Docker or Podman for consistent execution across dev, staging, and production.
- Multi-Cloud Management: Orchestrate cloud infrastructure across AWS, Azure, and GCP, optimizing for compute efficiency and cost.
Performance Tuning & Optimization
- GPU Optimization: Maximize hardware utilization for single-GPU and distributed multi-GPU environments.
- Algorithm Acceleration: Optimize Python code execution using Numba, NumPy, and specialized CUDA libraries.
- Load Testing: Conduct rigorous load and stress testing to guarantee system stability under high concurrent traffic.
Required Skills and Qualifications
Core Programming & Frameworks
- Language: Mastery of Python and its asynchronous ecosystem.
- ML Ecosystem: Deep expertise in PyTorch, Scikit-learn, and NumPy.
- Compilation: Experience accelerating Python code via Numba or Triton.
Conversational AI Experience
- Speech Technologies: Hands-on experience deploying Automated Speech Recognition (ASR) and Text-to-Speech (TTS) models.
- Generative AI: Familiarity with optimizing and serving Large Language Models (LLMs) and resource-efficient Small Language Models (SLMs).
Infrastructure & Operations
- Containers: Advanced knowledge of Docker, Podman, and container orchestration.
- Cloud Providers: Practical experience managing AI workloads on AWS (EC2, SageMaker), Azure (Azure ML), and GCP (Vertex AI).
- CI/CD Tools: Experience with GitLab CI, GitHub Actions, Jenkins, or specialized MLOps platforms (e.g., Kubeflow, MLflow).
Preferred Qualifications
- Conversational Context: Experience with dialogue management, prompt engineering, and Retrieval-Augmented Generation (RAG).
- Data Analytics: Familiarity with real-time data streaming (e.g., Kafka) and vector databases (e.g., Pinecone, Milvus, Qdrant).
- Quantization: Experience with model compression techniques like quantization (INT8/FP4), pruning, and distillation for edge deployment.