Petal Maps NLP Developer & Researcher

Responsibilities

1. Build multi-language NLP algorithm capabilities in map search and POI data production, including language recognition, word segmentation, NER, POI alignment, and domain map building.

2. Develop, implement, optimize, and deliver multi-language NLP algorithm requirements.

3. Large-scale geographic data analysis and processing, mining valuable information from data, and improving user search experience.

4. Continuously follow up the development of relevant technologies in academia and industry, and explore the application and implementation of industry-leading technologies in services.

Key Knowledge

1. Have an in-depth understanding of natural language processing, machine learning, and data mining algorithms, and be familiar with the principles and applications of classical NLP algorithm models.

2. Have solid engineering development and programming capabilities and be able to implement algorithm delivery.

3. Excellent problem-solving ability, learning ability, and teamwork awareness

Skills

1. At least one year of machine learning model development experience.

Nice to have

2. Familiar with NLP technologies, machine learning, deep learning algorithm, and deep learning framework platforms such as TensorFlow and Python.

3. Be familiar with Java, C++, and Python development languages and have good coding skills and habits.

Bonus items (Below items are also plus):

1) deep learning development experience

2) Experience in the development of search engines and information retrieval systems

3) Multi-language NLP algorithm experience

4) GIS service development experience

Detailed NLP Developer Short JD:

* Conduct necessary development work to deploy NLP models for production.

This may include;

• Java (spring/spring boot) development of ML/DL/NLP models in accordance with design patterns, from scratch or taking provided python code as reference.
• Setting up microservices.
• Close consultation with internal DevOps team.
• Providing feedback to the research team on deployability/feasibility of models given the available infrastructure.
• Optimizing models (compression/quantization etc) so that they can work within a given hardware/latency budget.