Atanas Yonkov
Software Engineer from Sofia, Bulgaria
Fullstack Engineer with 3+ years of experience building web applications for enterprise clients. Currently extending AI-powered agentic systems using LangChain and LangGraph at KPMG, while independently architecting a production-grade ERP system (React, FastAPI, PostgreSQL, Docker, Nginx) end-to-end. Master's from University of Amsterdam with Distinction, specializing in intelligent systems and HCI. Comfortable owning systems end-to-end, from schema design to deployment.
SKILLS
Languages
TypeScript, JavaScript, Python, SQL
Frontend
React, Next.js, Vue 2/3, TanStack Query, Redux, Vuex, Pinia, Tailwind CSS
Backend
Node.js, Express, FastAPI, Apollo GraphQL, REST APIs
AI/LLM
LangChain, LangGraph, AI Agents, Prompt Engineering
Databases
PostgreSQL, MongoDB, TSQL, SQLAlchemy, Prisma ORM, Knex.js
DevOps & Tools
Docker, Docker Compose, Nginx, Git, CI/CD, Webpack
Validation & Testing
Zod, Pydantic, React Hook Form, pytest
Concepts
OOP, Functional Programming, RESTful Design, GraphQL, Microservices
EXPERIENCE
Software Engineer at KPMG ITS
Sofia, Bulgaria
September 2024 – Present
Extend AI-powered agentic platform using LangChain and LangGraph that enables real-time client interaction with large-scale fiscal data (implemented Azure integration, multiprocessing for concurrent user sessions, and auto-scaling worker instances).
Build and extend internal Vue.js platform components serving multiple concurrent client engagements, improving feature delivery through reusable, modular architecture.
Mentor junior developers on codebase onboarding, PR review practices, and frontend architecture patterns.
Automate generation of data-driven reports (Excel, PowerPoint, PDF), eliminating manual workflows and ensuring cross-business-unit consistency.
Research Engineer at CWI & University of Amsterdam
Amsterdam, The Netherlands
February 2024 – June 2024 (5m)
Designed user study (n=16) with 7-test validation pipeline; built data collection system and delivered actionable product recommendations based on quantitative analysis.
Software Engineer at Merkle (Dentsu)
Sofia, Bulgaria
April 2022 – September 2023 (1y 6m)
Built environment management system adopted company-wide, enabling on-demand sandbox provisioning with automated dev/prod sync — reduced setup time by approximately 80%.
Developed internal employee rewards platform (Next.js, Apollo GraphQL, Knex.js), streamlining peer recognition and improving engagement.
Software Engineer Intern at GILO Technologies
Rijswijk, The Netherlands
July 2021 – February 2022 (8m)
Built core integration layer for Google Docs and Microsoft Word add-ons using TypeScript and Node.js.
Delivered production features for document automation used by end-customers for contract management.
PROJECTS
EDUCATION
Master of Science in Information Systems
University of Amsterdam
August 2023 – June 2024
Distinction (Cum Laude)
Specialized in intelligent systems, human-computer interaction, and data-driven product design.
Thesis research conducted at CWI (Dutch national research institute) — designed user study with 16 participants and built data analysis pipeline.
Bachelor of Science in ICT
Fontys University of Applied Sciences
August 2019 – August 2023
Erasmus exchange at Halmstad University (Sweden) — Computer Networks minor, Cisco CCNA fundamentals.
Built full-stack web applications across multiple team-based industry projects.
CERTIFICATES
AI Integrations for Developers
August 2025
SoftUni AI
- LangChain & LangGraph
- Vector Databases & Embeddings
- OpenAI & Anthropic API Integration
AI Tools for Developers
June 2025
SoftUni AI
- GitHub Copilot & Cursor AI
- Claude in Development
- Multi-Tool AI Integration