COMMUNITYAskapro

Multi-Document Summarization and Ranking Optimisation

  • Engineered a SOTA multi-document summarization model for news aggregation, achieving an 87% human acceptance rate with less than 1% critical errors.
  • Designed automated news timelines for evolving events, resulting in a 2.8% increase in content depth and a 1.4% boost in user time spent.
  • Optimized ranking algorithms with a CTR prediction model, increasing daily active users by 4.3% and user engagement by 2.1%.
  • Created a novel similarity scoring formula, improving F1-score from 0.91 to 0.95 for news clustering.
Andrey worked on this case as the ML engineer at Yandex Zen.
ML engineer
Natural Language Processing
Recommendation Systems
Entertainment
Media
Global
Enterprise
News & Content Platform
Web App
Python
Pandas
Hydra
TensorFlow
Scipy
Docker
Tensorboard
Wandb
B2C
Show more
Meber iconMeber thumbnail

Andrey

Machine Learning Engineer at Yandex Zen

Andrey's cases
Natural Language Processing
Large Language Models (LLM)
Benchmark Coverage Optimisation
Ranking Optimisation
Back Translation
AI
Artificial Intelligence
Large-Scale Text Corpus Deduplication and Dataset Enhancement
  • Developed and deployed text corpus deduplication using a suffix array algorithm on MapReduce, boosting assessor F1-score from 0.77 to 0.82.
  • Trained a classifier to improve benchmark coverage, enhancing dataset relevance for pretraining tasks.
  • Created a dataset augmentation pipeline with Back Translation, increasing pretraining robustness and generalization.
  • Enhanced document parsing quality, resulting in 3% faster model convergence and improved resource efficiency.
Language Translation Systems
Compiler Development
AST Parsing
Telecom
Global
Research and Development
C to Eolang Compiler
C to Eolang Compiler Development
  • Implemented processing of all basic data types from Clang AST to equivalent constructions in Eolang.
  • Developed a mechanism for translating multidimensional arrays to Eolang.
  • Developed a mechanism for translating Enum types to Eolang.
Search Algorithms
Logistics Optimization
E-Commerce
Russia
Product Search
Marketplace Platform
PostgreSQL
Product Search and Logistics Automation
  • Developed a search robot for products, increasing cold client conversions by 4%.
  • Created a report monitoring system for item positions for clients.
  • Developed an advanced route-planning algorithm for courier logistics, increasing daily pickup points by 13%.
  • Added generation of invoices for product stock, improving product acceptance process speed and accuracy by 15%.
Show more

Similar cases

Natural Language Processing
Recommendation Systems
Entertainment
Media
Global
Enterprise
News & Content Platform
Multi-Document Summarization and Ranking Optimisation
  • Engineered a SOTA multi-document summarization model for news aggregation, achieving an 87% human acceptance rate with less than 1% critical errors.
  • Designed automated news timelines for evolving events, resulting in a 2.8% increase in content depth and a 1.4% boost in user time spent.
  • Optimized ranking algorithms with a CTR prediction model, increasing daily active users by 4.3% and user engagement by 2.1%.
  • Created a novel similarity scoring formula, improving F1-score from 0.91 to 0.95 for news clustering.
Natural Language Processing
Large Language Models (LLM)
Benchmark Coverage Optimisation
Ranking Optimisation
Back Translation
AI
Artificial Intelligence
Large-Scale Text Corpus Deduplication and Dataset Enhancement
  • Developed and deployed text corpus deduplication using a suffix array algorithm on MapReduce, boosting assessor F1-score from 0.77 to 0.82.
  • Trained a classifier to improve benchmark coverage, enhancing dataset relevance for pretraining tasks.
  • Created a dataset augmentation pipeline with Back Translation, increasing pretraining robustness and generalization.
  • Enhanced document parsing quality, resulting in 3% faster model convergence and improved resource efficiency.
Demand forecasting
Recommendation Systems
NLP
Sentiment Analysis
Customer Churn Prediction
Fraud Detection
Business Optimization
AI/ML Solutions for Business Optimization
  • Developed AI/ML solutions to help businesses optimize processes and improve efficiency.
  • Implemented fraud detection systems to predict anomalies in transactions.
  • Developed demand forecasting models to optimize inventory and supply chains.
  • Created customer churn prediction models to analyze data and retain users.
  • Built recommendation systems to personalize content and products.
  • Conducted sentiment analysis using NLP to detect sentiments.
DevOps Architecture
Identity & Access Management
Kubernetes
Infrastructure as Code
EdTech
SaaS
Global
Scaling and Securing an EdTech Platform
  • Replaced legacy VM infrastructure with Kubernetes clusters, reducing infra costs by 30% and improving uptime.
  • Designed and rolled out a standardised IAM system using self-hosted Keycloak to decouple from third-party providers.
  • Managed distributed bare-metal NoSQL databases for fast-changing data, cutting storage costs by 50%.
  • Supported a rapidly growing codebase with 30+ sub-projects through modular cloud-native architecture.
CI/CD
MVP Development
Cloud Deployment
EdTech
Global
MVP
Adaptive Testing Platform
Full-Stack MVP for Adaptive Testing Platform
  • Built a functional MVP in 2 months using React, NestJS, and PostgreSQL.
  • Defined and deployed system architecture on AWS, including full CI/CD setup.
  • Wrote deployment documentation to support handover and onboarding.
  • Streamlined development, improving team productivity by 15%.
Show more