Enginering Software Development Machine Learning
About CADDi - Website (US Office): https://us.caddi.com/ - Youtube Video: https://www.youtube.com/@CADDi-Drawer_ Expected Role The Machine Learning Engineer will develop models in machine learning and data science, and build, maintain, and operate the infrastructure to continuously deliver these models to the service. The successful candidate will be expected to develop highly accurate modeling techniques that can leverage CADDi's data and provide value to our products, as well as develop stable systems in a team environment. Responsibility - Train, evaluate, and deploy sophisticated machine learning models to enhance the features of our drawing searching system. - Participate in the full software development cycle: design, develop, QA, deploy, experiment, analyze and iterate - Collaborate across disciplines and with other ML teams to find technical solutions to solve complex challenges - Monitor and continuously improve the deployed models. Examples of what success looks like in this role - Analyze drawing images and develop technology to extract information described on the drawings. - Construct image recognition models and annotation system - Study the use of large-scale language models (LLM) and large-scale visual models (LVM). - Collaborate with data science teams to bring prototype models to production - Raise the bar for engineering excellence and delivery
- Master's or PhD degree in ML, NLP, information retrieval, or a related field. - ~5 to 7 years of proven experience as a Machine Learning Engineer or in a similar role, with a strong track record of developing and deploying machine learning solutions in real-world applications within a cloud environment. - Solid knowledge of algorithms related to machine learning, statistics, linear algebra, and computer science - Experience in image recognition, OCR, and 3D analysis - Experience with GPU-based data processing (CUDA, OpenCL, cudf, CuPy, etc.) - Experience developing machine learning pipelines using Vertex AI Pipeline, kubeflow, Apache Beam, Spark, etc. - Experience working with machine learning to solve business problems - Experience improving the accuracy of machine learning and statistical models - Proven track record of deploying machine learning models to production environments - Experience working with cloud services such as Google Cloud and AWS - Business communication skills in English (for daily communications and meetings) PRODUCT DEVELOPING ENVIRONMENT Languages: - Front-end: TypeScript - Backend: Rust, TypeScript, Python Framework/ Library - Frontend: React, Next.js, WebGL, WebAssembly - Backend: Rust (axum), Node.js (Express, Fastify, NestJS), PyTorch Infrastructure: Google Cloud, Google Kubernetes Engine, Anthos Service Mesh Database/Data Warehouse: CloudSQL (PostgreSQL), AlloyDB, Firestore, BigQuery APIs: GraphQL, REST, gRPC Monitoring: Datadog, Sentry, Cloud Monitoring Environment construction: Terraform CI/CD: Github Actions Authentication: Auth0 Development tools: GitHub, GitHub Copilot, Figma, Storybook Communication tools: Slack, Discord, JIRA, Miro, Confluence
Working style: - Hybrid (come to Office at least once a week) Holidays and leave: - Annual paid leave: 12 days - National holidays - Year-end holidays (December 31 to January 2) - Tet holidays - Others (following Labor Regulations) Benefits: - Salary review every 6 months - Employee Stock-option - 13th month salary - Salary review: twice a year - 100% monthly basic salary and mandatory social insurances in 2-month probation - Premium Health Insurance - Social insurance, health insurance, unemployment insurance, workers’ accident compensation insurance - Annual health check-up - Intensive training program (external or internal training courses, workshop etc) - Devices: PC and display of desired specifications - Awards: Company awards, every 6 month MVP awards - Activities: Year-end-party, team building, etc
- Experience in development and operation related to distributed processing - Business level Japanese proficiency
EM
HackerRank test (300min) -> Casual Interview -> Technical interview (could be combined) -> Final interview (with CTO) -> Offer interview