AI/ML Software Engineer

At Huex, we are not just building products; we are building a culture of continuous learning, creativity, and empowerment. We want to inspire you to think differently, embrace new technologies, and push the limits of what is possible. Join us on this exciting journey, and together we will shape the future of the digital landscape

Requirements

• Bachelor’s degree in Computer Science, Artificial Intelligence, Machine Learning or a closely related field
• 2–3 years of hands-on experience developing, fine-tuning and deploying AI/ML models using Python and frameworks such as TensorFlow, PyTorch or Keras
• Proven track record working with sparse vector representations for high-dimensional data, especially in NLP pipelines and recommendation engines
• Practical expertise integrating AI models with IoT architectures, including MQTT and CoAP protocols, AWS IoT Core or Azure IoT Hub, and edge devices like NVIDIA Jetson or Raspberry Pi
• Experience designing real-time inferencing pipelines for large-scale sensor or device streams, ensuring sub-100 ms latency and high throughput with tools such as ONNX Runtime, TensorRT or Triton
• Strong background in embedding LLMs into healthcare workflows: extracting EHR summaries, performing clinical note classification, generating diagnostic suggestions, and applying HIPAA-aware data handling (FHIR/HL7)
• Hands-on with LLM fine-tuning and prompt engineering using Hugging Face Transformers, LangChain or similar toolchains, plus experiment tracking via MLflow or Weights & Biases
• Familiarity deploying and monitoring models in production with MLOps platforms like Kubernetes, Docker, Kubeflow or Airflow, and observability tooling (Prometheus, Grafana)
• Understanding of register pressure, shared-memory usage and GPU optimization—profiling with Nsight Systems/CUDA Toolkit and eliminating compute or memory bottlenecks
• Comfortable architecting for both memory-bound and compute-bound workloads, leveraging quantization, pruning and mixed-precision training to maximize performance on resource-constrained devices
• Solid skills in version control (Git) and collaborative development on GitHub or GitLab, including pull-request workflows, code reviews and CI/CD pipelines
• Familiarity building AI-enhanced CRM features—automated lead scoring, follow-up orchestration and conversational interfaces—integrating with REST/GraphQL APIs and webhook-driven architectures
• Strong problem-solving aptitude for tuning model performance, scalability and cost-efficiency, with experience using distributed training (Horovod, DeepSpeed) when required
• Awareness of security best practices for IoT and healthcare systems, including secure boot, certificate management and encrypted data in transit and at rest
• Excellent communication skills to collaborate with cross-functional teams, translate technical constraints into product requirements and drive innovation at the intersection of AI, IoT and digital health

About the Role

• Design and build AI and ML models that use sparse vector representations to efficiently process high-dimensional data from sources like EHRs and sensor streams
• Implement and fine-tune deep learning architectures for NLP tasks (EHR summary extraction, diagnosis suggestions), computer vision use cases or time-series analytics on IoT devices such as pet wellness trackers
• Profile and optimize both memory-bound and compute-bound operations, reasoning about register pressure, shared-memory usage and GPU utilization with tools like NVIDIA Nsight and CUDA profilers
• Apply the latest techniques—quantization, pruning, mixed-precision, model distillation—to accelerate inference and training workloads on resource-constrained devices and edge gateways
• Integrate AI pipelines with IoT ecosystems via MQTT, CoAP or HTTP/REST, enabling real-time data ingestion, decision-making at the edge or cloud orchestration through platforms like AWS IoT or Azure IoT Hub
• Develop and deploy real-time inferencing services using ONNX Runtime, TensorRT or Triton Inference Server to handle large-scale device telemetry with minimal latency
• Embed and optimize large language models—fine-tuning with Hugging Face Transformers or LangChain workflows—for tasks like clinical note classification, personalized chatbot support or CRM lead scoring and follow-up automation
• Collaborate with firmware, backend and data-engineering teams to seamlessly integrate AI components into broader product architectures and CI/CD pipelines
• Monitor production models with MLOps frameworks (Kubeflow, MLflow, Weights & Biases) and observability stacks (Prometheus, Grafana), continuously retraining and tuning for robustness and reliability
• Keep abreast of emerging AI, ML and IoT trends—edge inference accelerators, federated learning, new sparse-representation methods—to drive ongoing innovation and maintain a competitive edg

You have to wait 20 seconds

Generating Apply Link...