Senior Software Engineer

@Finonex

Designed, developed, and maintained a distributed lead recommendation and scoring system for fintech call centers.

• •Built reactive lead scoring system using Python, Redis, Lua, and Kafka to process 3M+ leads in real-time across thousands of sales representatives.
• •Developed call dialog synchronization and A/B testing features for lead conversion optimization.
• •Migrated deployments from raw YAML to Helm for improved maintainability.
• •Enhanced CI/CD pipelines with automated integration testing and ephemeral environment provisioning in Kubernetes.
• •Designed APIs with Express, MariaDB, Redis, and Kafka.
• •Built desktop and web apps for call center agents and support teams.
• •Reduced AWS RDS costs through data modeling and query optimization.
• •Developed monitoring dashboards and alerts in Grafana.
• •Contributed significantly to the development of an in-house CRM system.

Senior Software Engineer

@Fieldin

Worked in DevOps R&D on automation for smart farming systems.

• •Designed distributed automation and regression testing platform using Python, RabbitMQ, Redis, and AWS.
• •Replaced 6 static QA environments with ephemeral test setups—cutting costs drastically and accelerating development.
• •Developed real-time queue message replication service in Go.
• •Built Angular web interface for automation orchestration.
• •Enhanced Jenkins CI/CD pipelines for infrastructure and software workflows.

Software Engineer

@Bosch Bulgaria

Worked on a car interior monitoring product for the automotive industry.

• •Contributed to the development of a distributed model evaluation toolchain using Python, RabbitMQ, Protobuf.
• •Processed model predictions from computer vision teams and labeled data from data engineering teams in various formats (CSV, SQL, NoSQL, JSON, XML, ADTF).
• •Generated videos with visualised model predictions, testing data and errors for each frame using OpenCV.
• •Computed model evaluation metrics for regression and classification models using Numpy, Pandas, Scikit.
• •Visualised model metrics and KPIs using Matplotlib, Seaborn.
• •Developed automation scripts for data processing.
• •Developed CI jobs for different KPIs using Jenkins.
• •Contributed to the development of a pose estimation model based on OpenPose paper, trained on COCO dataset using PyTorch.

Machine Learning Researcher

@GATE Institute

Worked in collaboration with Chalmers University of Technology.

• •Research and development on building detection, roof recognition, and geometry generation for Automatic 3D LOD2 Building Model Reconstruction.
• •Research and development on roof recognition — roof type classification and roof plane semantic segmentation from orthophotos and LiDAR point clouds using Deep Learning approaches with PyTorch.
• •Published a scientific paper for "IFAC Workshop Control for Smart Cities – CSC 2022". The topic of our paper was "Roof Segmentation Towards Digital Twin Generation in LoD2+ Using Deep Learning".
• •Wrapped C++ core software of Digital Twin Cities Centre as an API with Node.js Addons.
• •Developed a web-based 3D City Viewer for CityJSON and CityModel (DTCC internal) formats with Babylon.js.

• •Contributed to the design, development, maintenance, and management of over 20 web solutions for b2b and b2c businesses using Laravel, Express, Vue, MySQL, MongoDB, BrainTree.
• •Established working relationships with 4 companies in the digital industry.
• •Supported 5 tech events through sponsorship.

Designed the distributed system and infrastructure/deployments. First (and so far only) customer is the largest e-commerce in Bulgaria; onboarding of a second is in progress. Still pre-launch as an official SaaS. Now mostly consulting — limiting my coding and DevOps load on the project.

Solo freelance build for Bulgaria's State Reserve. Security personnel at national bases use it to register visitors (personal details, ID documents, vehicle info, purpose of visit), track entry/exit timestamps, search historical records, and print visitor permits. We also installed ID-scanner hardware on-site. The system has been in daily use for two years.

Co-authored research on roof type classification and roof-plane semantic segmentation for Automatic 3D LoD2+ Building Model Reconstruction. A collaboration between GATE Institute (Bulgaria) and Chalmers University of Technology (Sweden). Published in IFAC-PapersOnLine under the title “Roof Segmentation Towards Digital Twin Generation in LoD2+ Using Deep Learning”.

Built for my brother's company, which designs and manufactures vacuum soldering machines end-to-end. I handled the web side.

The first team-format grappling event of its kind in the country — organized and co-sponsored the inaugural edition.

Contributed to the campaign for the first custom-designed playing cards from Bulgaria. Pledged $28,990 against a $6,400 goal on Kickstarter.

Ran across 3 Bulgarian cities. Goal: motivate young people toward software and digital careers early.