Ivan Peev, Developer in Sofia, Bulgaria
Ivan is currently unavailable

Ivan Peev

Bio

Ivan is a full-stack developer with 20+ years of experience programming and leading development teams for projects in various industries. He is keen on working with Python, C++, Java, SQL, and both Linux and Windows platforms. Throughout his career, Ivan has handled mathematics, statistics, numerical methods, computational geometry databases, data-collecting bots, and principles for developing software applications.

Portfolio

KPMG
Python, Vertex AI, LangChain, LangGraph, BigQuery, Cloud Run, Azure DevOps...
Hexagon
C++, Python, Docker, REST APIs, Jupyter Notebook, Mathematics...
Yields NV
Python 3, Python, Git, Lint, Machine Learning, Spark, Hadoop, SciPy, NumPy...

Experience

  • C++ - 20 years
  • Python - 12 years
  • Oracle - 10 years
  • Linux - 10 years
  • Java - 10 years
  • Numerical Methods - 10 years
  • Statistics - 5 years
  • PostgreSQL - 4 years

Preferred Environment

Windows, Linux

The most amazing...

...project I've developed is a Python agent that automatically upgraded server environments using search algorithms.

Work Experience

Senior Software Developer | AI Engineer

2025 - 2025
KPMG
  • Served as a core contributor to KPMG’s Agent Framework, developing production AI agents on GCP for audit and advisory workflows.
  • Orchestrated multi-step agents using LangChain, LangGraph, Google ADK, and Vertex AI, integrating with BigQuery and Cloud Storage for enhanced functionality.
  • Implemented human-in-the-loop review/approval systems, guardrails, and evaluations using Vertex AI Evaluation, with logging and observability via cloud logging and monitoring.
  • Delivered Python/FastAPI services on Cloud Run, utilizing CI/CD with Azure DevOps (repos/pipelines) and Cloud Build, and containerized the applications with Docker.
Technologies: Python, Vertex AI, LangChain, LangGraph, BigQuery, Cloud Run, Azure DevOps, FastAPI, Docker, Retrieval-augmented Generation (RAG), Large Language Models (LLMs), AI Agents

Senior C++/Python Developer

2024 - 2025
Hexagon
  • Developed 2D/3D geometry analysis library for CMM systems, focused on precision inspection of curves and surfaces.
  • Implemented and maintained algorithms for point cloud analysis from contact and optical sensors.
  • Created REST APIs for integrating geometry software for 2D/3D analysis of industrial components.
  • Refactored the C++ back end and added Python modules for testing, data processing, scripting, and visualization.
Technologies: C++, Python, Docker, REST APIs, Jupyter Notebook, Mathematics, Computational Geometry, Visual Studio, Simulations

Python Developer

2023 - 2023
Yields NV
  • Developed a product for monitoring the stability and quality of complex mathematical models.
  • Integrated the BNP Paribas bank's model risk process into the Yields product.
  • Wrote Python library, Jupyter notebooks, and user documentation. Used GitHub and Jira.
Technologies: Python 3, Python, Git, Lint, Machine Learning, Spark, Hadoop, SciPy, NumPy, Pandas, Jupyter, Jupyter Notebook, Credit Modeling, GitHub, Pytest, Microsoft Excel, Kubernetes, Back-end Development

Software Developer

2022 - 2023
TRM Labs Inc.
  • Created ETL pipelines that ingest, transform, and process large amounts of structured and unstructured data in real time, ultimately helping financial institutions and governments fight cryptocurrency fraud and criminal activity.
  • Built highly reliable continuous crawling pipelines that collect data from the open web, forums, etc.
  • Designed systems, patterns, and processes for ingesting new data from the open and dark web used by teams across TRM.
Technologies: SQL, Python, ETL, Web Scraping, Big Data, Google Cloud, Google BigQuery, Apache Airflow, Visual Studio Code (VS Code), Pandas, GitHub, JavaScript, PySpark, REST APIs, Pytest, Blockchain, Google Cloud Platform (GCP), TOR, Data Analysis, Docker, Kubernetes, Crypto, BigQuery, Back-end Development

Senior Back-end Developer

2022 - 2022
Wines of Kings LLC
  • Served as an API integration developer. Created a mini Bloomberg with wines to constantly compare portfolio with the market: compare the product list price in Shopify Store against the competition to check if it is always competitive.
  • Contributed to the app that tracks if there is any uniquely interesting offer from our European sources on any wine currently carrying.
  • Used Shopify API, wine-searcher API, Python, Django, Pandas, Airtable API, Scraping API, PostgreSQL, Linux, and Jenkins.
Technologies: Python, APIs, Shopify API, Web Scraping, PostgreSQL, Pandas, Airtable, Jenkins, Django, Django ORM, REST APIs, Microsoft Excel, Docker, Docker Compose, Back-end

Software Developer

2016 - 2021
Group Numerics
  • Developed a connection risk estimator software for SafeConnect. Online travel agencies can integrate the risk API to rate connecting flights in real time, and users can rely on those ratings to book suitable flight options.
  • Developed small projects and algorithms with Linux, Python 3.6 and 3.7, Airflow, Pandas, Jupyter Notebooks, PostgreSQL, and AWS Lambda.
  • Developed a control machine details software for Hexagon. It is a unique numerical software for the airfoil of blades to coordinate measuring machines (CMM) using C++ and Python.
Technologies: Python, PostgreSQL, C++, Numerical Methods, Numerical Analysis, Computational Geometry, Mathematics, Statistics, Django REST Framework, Pandas, Apache Airflow, Redis, AWS Lambda, JavaScript, C, Amazon Web Services (AWS), Docker, REST APIs, Linear Algebra, Docker Compose, Back-end, FastAPI, Architecture

Java Team Leader

2011 - 2016
Playtech
  • Managed a server development team of about 12 members, including Java, SQL, and PHP developers. The team developed the majority of Playtech casino games.
  • Wrote simulations to test the probabilities of games in multiple ways. Developed the central system in Java and connected it to Oracle databases running in a cluster of connected servers.
  • Created an API for Flash, Windows, and mobile clients. Used Java, Flash, and PHP to develop the web interface for the administrative back end.
Technologies: Java, Oracle, Python, Linux, Probability Theory, Mathematics, Statistics, Leadership, Spring Boot, Microsoft Excel, Data Analysis, Spring, Back-end, Apache Kafka, Back-end Development, Architecture, Simulations

C++ Team Leader

2009 - 2011
Travelstoremaker
  • Handled the development of the Wires project, which involved programming the main company product—a reservation system for travel agencies that allows booking flights, cars, and hotels.
  • Developed multiple connections to external reservation systems (GDS) and other reservation service providers, mainly using web services.
  • Used C++ and an Oracle database to develop the system. The web interface worked with common gateway interfaces (CGI) written in C++ and PHP with rich, dynamic HTML.
Technologies: C++, Oracle, Linux, Python, Leadership, Back-end, Back-end Development

Embedded Developer

2005 - 2008
Johnson Controls
  • Wrote automotive software for the auto industry. Developed a custom multifunction display embedding real-time operating systems for Peugeot 307 and 407 and Citroen C4 and Xsara.
  • Used C and Python for software development. Worked with ARM7 and Fujitsu M32C processor architectures. Created both color and black and white displays.
  • Worked on all levels of system development, following an Automotive SPICE-based process. Embedded the Windows System, Automotive Controller Area Network (CAN), HAL and drivers, clock precision, and temperature compensation.
Technologies: Python, C, Embedded C, Embedded Software, AUTOSAR, Automotive Controller Area Network (CAN), CANopen, ARM Architecture, Embedded Systems, Embedded C++, CAN Bus

Development Manager and Software Developer

2001 - 2004
i:Fao
  • Managed the cytric v7 project that continuously evolved the primary company product, a travel reservation system. Led a team of 15 Java developers and one web designer.
  • Created travel management software for large and middle-size corporations to manage flight, rail, car, and hotel reservations. It had multiple connections to external reservation systems (GDS) and other providers, mainly using web services.
  • Designed the system architecture using the Oracle Database and WebLogic Application Server on Linux. The application had a web interface with rich, dynamic HTML.
Technologies: Java, Oracle, Leadership, Back-end, Architecture

Software Developer and Customer Support Manager

1997 - 2000
Rodopi Software
  • Developed billing and management software for internet providers in the scope of the Rodopi project. It involved integrating credit card processing software, automated internet subscription, and web hosting.
  • Ensured the development of VoIP billing, RADIUS Server billing, net and web traffic collectors integration, and mail and RADIUS servers automated provisioning.
  • Used C++, Transact-SQL, Microsoft SQL Server, IIS, ASP, CGI, COM, and Perl for the project development.
Technologies: C++, Microsoft SQL Server, SQL

Experience

Connection Risk Estimate Software for SafeConnect

A risk API to be integrated by an online travel agency to rate connecting flights in real time. The ratings help users book the most suitable flight options.

My contributions to this project involved developing the risk API as a Python Django web application and a PostgreSQL database.

Spline Curve Calculations Library

The Curve library is a numeric library that performs spline curve calculations on a point data set. The splines constructed are interpolated and constrain third-order smooth splines. The library is specially designed to be suitable for evaluating real data sets obtained with instruments or methods that may have a high level of uncertainty. Due to that reason, the data need to be processed accurately, considering the possibility of having poorly distributed data.

Competitive Benchmarking for Airlines and Online Travel Agencies

Developed software in Python, PostgreSQL, and Pandas to create over 40 unique report formats, offering airlines and online travel agencies the ability to understand their competitive position route by route. Benchmarked various dates, origins and destinations, markets, and currencies. Ensured the unique fare benchmarking of each airline itinerary offered by online travel agencies to know strictly at what price your flights are being sold and by whom. Conducted one-shot campaigns or continuous benchmarking.

Education

1990 - 1995

Master's Degree in Computer Science

Wroclaw University of Science and Technology - Wroclaw, Poland

Skills

Libraries/APIs

REST APIs, Pandas, PySpark, Shopify API, Django ORM, SciPy, NumPy

Tools

Pytest, GitHub, Microsoft Excel, Docker Compose, BigQuery, Jenkins, Apache Airflow, Git, Jupyter, Visual Studio

Languages

Python, Java, C++, C, SQL, Embedded C, JavaScript, Embedded C++, Python 3

Platforms

Windows, Linux, Oracle, Jupyter Notebook, AWS Lambda, Visual Studio Code (VS Code), Amazon Web Services (AWS), Blockchain, Google Cloud Platform (GCP), Kubernetes, Apache Kafka, Docker, TOR, Vertex AI, Cloud Run

Frameworks

Django, Django REST Framework, Spring Boot, Spring, Spark, Hadoop, LangGraph

Paradigms

REST, ETL, Azure DevOps

Storage

PostgreSQL, Microsoft SQL Server, Redis, Google Cloud

Other

Software Development, Numerical Methods, Back-end, Back-end Development, Network Architecture, Numerical Analysis, Probability Theory, Computational Geometry, Mathematics, Statistics, Embedded Software, Leadership, AUTOSAR, Embedded Systems, Linear Algebra, CAN Bus, FastAPI, Architecture, Simulations, APIs, Web Scraping, Airtable, Big Data, Google BigQuery, Automotive Controller Area Network (CAN), CANopen, ARM Architecture, Lint, Machine Learning, Credit Modeling, Coordinate Measuring Machine (CMM), Data Analysis, Crypto, LangChain, Retrieval-augmented Generation (RAG), Large Language Models (LLMs), AI Agents

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring