Description
Our team is focused on improving the process of concluding both standard and complex-structured investment banking deals for the Investment Activity Department using artificial intelligence.
Our customers are employees of various units and roles within the Investment Activity Department.
Responsibilities
What you will be doing:
- Review of modern global trends in artificial intelligence, prototyping
- Analysis of opportunities to improve customer products and processes using artificial intelligence
- Direct participation in projects to implement artificial intelligence into customer processes:
- Analysis of availability and applicability of existing technologies within the bank's perimeter
- Development and implementation of solutions using artificial intelligence
- Code development and optimization of parameters for used models
- Interaction with DE, integration of models into the computational process
- Development of LLMs for QA and Information Extraction tasks
- Implementation of data preparation and slicing pipelines, as well as prompt selection
- Organization and automation of the labeling process (from data search to quality checks via crowdsource labeling)
- Releases of new models into execution environments for our users.
Requirements
What is important for us:
- Mandatory: Completed higher education from key universities: Moscow State University / Moscow Institute of Physics and Technology / Russian Economic School / Higher School of Economics / National Research Nuclear University MEPhI / Bauman Moscow State Technical University, Moscow Aviation Institute
- Experience as a Data Scientist for 3+ years in projects for development and/or modification or implementation of software using artificial intelligence
- Excellent knowledge of Python: PyTorch, NumPy Sklearn, Pandas, experience with libraries Scipy, Matplotlib, Spark, Pyspark Hive
NLP
- Understanding of the main methods and algorithms of Machine Learning, knowledge of the principles of neural network models for text processing, experience in ranking recommendations NLP tasks
- Algorithmic training knowledge of main algorithms and data structures
- Experience in classic ML development of one or several types of models: Text Classification, NER, QA
- MLOps, Git Docker, MLFlow, DVC, ClearML, Airflow, experience in launching NLP projects from labeling to "production"
- Good knowledge of algorithms and data structures
LLM
- Experience with LLM Prompt Engineering, fine-tuning GPT-Like models, LLMOps, LangChain, LlamaIndex, experience with Plugins for LLM
- Desire to study new approaches models and technologies
- Advanced level in SQL
Will be a plus:
• Good GitHub profile
• Participation in open-source projects with LLM
• Participation and positive results in hackathons.
Conditions
What we offer:
- Official employment
- Comfortable office near Leninsky Prospekt metro station
- Annual salary review and annual bonus
- Corporate gym and relaxation areas
- Unique Sber training system for professional and career development
- Extended voluntary health insurance and preferential insurance for family
- Flexible discount on mortgage credit, equal to 1/3 of the Central Bank's key rate
- Free SberPrime+ subscription, discounts on products from partner companies
- Reward for referring friends to the Sber team
- Corporate pension program.
