Nikolay

Nikolay Dimitrov
Senior AI/ML Engineer
[відкрити контакти](див. вище в блоці «контактна інформація») [відкрити контакти](див. вище в блоці «контактна інформація») Sofia, Bulgaria

Summary

Senior AI/ML Engineer with 10+ years of experience building data-driven platforms, machine learning systems, and scalable
AIpowered applications. Experienced in developing production ML pipelines, large-scale data processing systems, and deploying
models using modern MLOps practices. Strong background in Python, deep learning frameworks, cloud infrastructure, and distributed
systems. Proven track record designing intelligent analytics platforms, optimizing model performance, and delivering reliable AI
solutions in cloud-native environments.

Skills

Programming Languages Machine Learning & AI
Python, JavaScript, TypeScript, Java, SQL PyTorch, TensorFlow, Scikit-learn, XGBoost, Hugging Face Transformers,
LangChain, LLM APIs, NLP, Deep Learning, Computer Vision
Data Engineering
Apache Spark, Pandas, NumPy, Airflow, Kafka, Feature MLOps & Infrastructure
Engineering, Data Pipelines, ETL/ELT MLflow, Kubeflow, Docker, Kubernetes, CI/CD for ML, Model Monitoring, Feature
Stores, Experiment Tracking
Cloud Platforms
AWS (SageMaker, Lambda, ECS, EKS, S3, RDS), GCP Architecture
(Vertex AI, BigQuery), Azure ML Distributed Systems, Microservices, Event-Driven Architecture, Data Lakehouse
Architecture
Databases
PostgreSQL, MongoDB, Redis, DynamoDB, Elasticsearch, Tools
Data Warehouses GitHub Actions, GitLab CI/CD, Terraform, Prometheus, Grafana, Jupyter, Weights &
Biases

Professional Experience

Senior AI/ML Engineer, Vector Software Ltd 03/2020–03/2026
Oslo, Norway
Led development of an AI-driven analytics platform built on Lakehouse architecture that analyzed large-scale
(Remote)
enterprise datasets and generated predictive insights for business operations.
Designed distributed ML pipelines using Apache Spark and Python, enabling large-scale feature engineering and
model training across billions of data records.
Implemented forecasting models (XGBoost, Prophet) to predict usage trends and operational KPIs, improving
forecasting accuracy by 35%.
Developed an automated data quality monitoring system using anomaly detection models that identified irregular
patterns in data pipelines.
Built scheduled validation pipelines with FastAPI, Celery, and Python ML models, automatically detecting schema
changes and data drift across ingestion pipelines.
Created alerting and monitoring dashboards integrated with Prometheus and Slack, reducing incident response time
for data issues.
Built an internal semantic search engine using transformer-based NLP models (BERT/Hugging Face) to index
engineering documentation and internal knowledge bases.
Implemented vector embeddings and similarity search using Elasticsearch and vector databases, improving internal
information retrieval across engineering teams.
Integrated LLM-powered summarization tools that automatically generated insights from large technical documents.
Developed recommendation models that suggested relevant datasets, dashboards, and analytics workflows for users
of the platform.
Built ranking algorithms using gradient boosting models and collaborative filtering to improve content discovery.
Implemented real-time recommendation APIs deployed via Docker and Kubernetes on AWS EKS.
Designed and implemented MLOps workflows for training, tracking, and deploying models using MLflow and
Kubernetes.
Built CI/CD pipelines that automated model testing, versioning, and deployment into production microservices.
Developed model monitoring services tracking model drift, prediction accuracy, and data quality metrics.
Machine Learning Engineer, DevOcean Solutions 09/2016-02/2020
Sofia, Bulgaria
Developed machine learning models that analyzed enterprise operational data to generate predictive insights for (On-Site)
planning and resource allocation.
Built regression and classification models using Scikit-learn and XGBoost to forecast project timelines and operational
performance.
Implemented automated pipelines for data ingestion, feature engineering, and model retraining using Python and
PostgreSQL.
Designed anomaly detection models that identified suspicious financial transactions and abnormal activity patterns.
Implemented unsupervised ML models such as Isolation Forest and clustering algorithms to flag high-risk activities in
real time.
Integrated ML predictions into backend APIs used by enterprise dashboards.
Developed ML-powered reporting services that analyzed operational datasets and automatically generated reports
and insights.
Built Python services that extracted patterns from large datasets and delivered predictive insights through dashboards.
Built scalable model inference services using FastAPI and Docker, allowing business applications to request
predictions in real time.
Implemented Redis caching and message queues to support high-volume prediction workloads.

Software Engineer (Data & AI Projects), Devision 09/2014–08/2016
Sofia, Bulgaria
Developed data processing pipelines that analyzed user interaction data from web platforms to generate insights on
(On-Site)
engagement and usage patterns.
Built data aggregation services using Node.js and Python to support analytics dashboards.
Implemented a recommendation prototype that suggested relevant content for users based on behavioral signals.
Built collaborative filtering algorithms and ranking models using Python.
Developed backend services that processed operational data from enterprise platforms and generated analytics
reports.
Built SQL-based ETL pipelines that structured large datasets for business intelligence tools.

Education

Technical University of Sofia 2012–2014
Master's Degree - E-Management
Technical University of Sofia 2007 – 2013
Bachelor's Degree - Computer Science
Technical School ELSYS, Sofia 2002 – 2007
IT Specialist - Computer Software Engineering

Languages

English Bulgarian
German