Ян

YM

YAN MARCHAN
[open contact info](look above in the "contact info" section) | [open contact info](look above in the "contact info" section) | Vilnius | WWW: [open contact info](look above in the "contact info" section)

Summary

Innovative Python Developer known for high productivity and efficient task completion. Possess specialized skills in data
analysis, machine learning algorithms, and web development frameworks. Excel in problem-solving, communication, and
teamwork, ensuring projects are delivered with precision and collaborative effort.

Skills

• Programming languages • Monitoring tools
• API development • Project management
• Database management • Documentation tools
• Container orchestration
• Microservices architecture

Experience

Aspen Technology Labs | Remote Built as part of a multi-engineer team, this project is an AI-assisted job
Python Developer scraping and feed-management platform that combines a FastAPI backend,
09/2025 - Current async Python workers, Playwright-based scraping, LLM-powered pattern
extraction, PostgreSQL/ClickHouse, AWS SQS/S3, SQL migrations, and a
React operations UI to discover career pages, scrape job data, manage scraping
requests/configurations, and export results.
• Developed backend API and orchestrated worker processes, focusing on
scraping_requests, jobs_data, queue-worker scheduling, and mini-
spider/feed processing flows.
• Implemented and refined scraping request workflows, including domain-
based and job-page-URL-based request modes, request-scoped processing,
rescrape behavior, and stricter domain/job-page URL validation.
• Enhanced filtering, sorting, partial-update, and export capabilities for
scraping requests and jobs data endpoints, improving data accessibility.
• Improved reliability and recovery for stuck or timed-out feeds, addressing
stale processing states, draining queue backlogs, and implementing auto-
removal scheduling.
• Added integrations and supporting changes across API
schemas/services/repos, SQL patches, S3-backed file indexing/export flows,
and worker-side processing.

Bachelor's project | Vilnius • Built an AI-assisted scraping pipeline that converts natural-language
Python Developer prompts into structured extraction logic using LLM-based intent analysis,
08/2025 - 02/2026 schema extraction, and regex generation/refinement.
• Designed and developed a microservice web data extraction platform
using FastAPI, Celery, Redis, PostgreSQL, and Docker Compose.
• Implemented a Playwright headless worker to scrape dynamic websites,
capture semantic page content, full HTML, and network responses for
downstream processing.
 • Enhanced parser caching and normalized database design (3NF) to
facilitate extraction reuse, minimize repeated LLM calls, and streamline
task/result storage.
• Implemented secure API features such as JWT authentication, rate limiting,
input sanitization, correlation IDs, and health-check endpoints to safeguard
data and improve service reliability.
• Created support for scheduled scraping jobs and end-to-end async task
orchestration across API, workers, and scheduler services.
• Built a comprehensive automated test suite with 200+ tests and roughly
75%+ coverage, including unit, integration, contract, edge-case, and
performance tests.
• Created comprehensive technical documentation, OpenAPI docs,
architecture diagrams, and deployment artifacts to support development
and ensure project clarity.

Aspen Technology Labs | Remote • Optimized Velocity scripting | Delivered client-focused support
Web-Scraping Engineer • Facilitated data extraction for numerous enterprise clients, including
12/2023 - 09/2025 prestigious accounts with over 10,000 unique web sources, ensuring
consistent and scalable performance.
• Engineered and refined Velocity-based parsing logic within platform
constraints, achieving clean and structured data output with minimal
resource consumption.
• Optimized configurations and queries, reducing latency and increasing
throughput for high-volume clients.
• Collaborated with clients and technical stakeholders to gather
requirements, resolve data discrepancies, and align on project expectations.
• Acted as a liaison between clients and internal teams, translating technical
challenges into actionable solutions.
• Diagnosed and resolved complex scraping and parsing challenges across
diverse site architectures, proactively safeguarding data integrity.
• Adopted a solution-oriented approach in both technical configurations and
client interactions, fostering improved performance and building trust.
• Enhanced data extraction capabilities, enabling large enterprises to
efficiently access and utilize web data.

EHU IT Hub | Vilnius • Developed a Telegram group access-control bot that verified users via
Python Developer, Part-time university email and automatically restricted or removed unverified
04/2024 - 01/2025 members from protected chats.
Implemented the full backend flow using Python, Aiogram, SQLAlchemy,
PostgreSQL, and Redis, including user/group persistence, verification state
management, SMTP-based code delivery, and scheduled cleanup jobs.
• Built admin/group management features, including bot permission checks,
ownership tracking, group status management, and restoration of chat
permissions post-verification.
• Added multilingual support, executed database migrations, created
automated tests for models/services/utilities, and containerized
deployment using Docker, Helm values, and GitLab CI/CD pipelines.

Education

School of Digital Competencies (EHU) | Vilnius, Lithuania
Bachelor of Science in Computer Science
02/2026
Taras Shevchenko National University of Kyiv | Kyiv, Ukraine
Data Analyst in Computer Science

Websites, Portfolios, Profiles

• http://github.com/marchanyanehu/

Projects

• Chat Guard Telegram bot, aiogram, postgresql, Developed Chat Guard Telegram bot that monitors chats and checks
for unverified users.
• CRM for EHU IT hub, FastAPI, postgresql, Developing CRM for automation of the flow of applicants to the EHU IT
hub.
• Microservice architectural solutions, FastAPI, postgresql, Designing and developing microservice architectural
solutions to integrate into all companies' projects.
• ETL pipeline, Designed and implemented the ETL pipeline to extract, transform, and load data from diverse sources,
using custom scripts to handle complex data transformation logic.
• Data models using SQLAlchemy ORM, Created multiple data models using SQLAlchemy ORM to represent and
manipulate entities such as ItemSimple, ItemDetailed, and ItemProcessed.
• API clients for real-time data, FastAPI, Developed API clients to fetch real-time data from external APIs, utilizing the
FastAPI framework for handling HTTP requests.
• Database schema changes using Alembic, Managed database schema changes using Alembic for smooth data
migrations and version control.
• Machine learning models using XGBoost, XGBoost, Built machine learning models using XGBoost to predict specific
outcomes, such as price fluctuations or item demand, achieving high accuracy and performance metrics (MSE and R²
score).
• Utility scripts for data validation and visualization, Matplotlib, Integrated utility scripts for data validation and
visualization, leveraging Matplotlib for detailed performance analysis.