Roman Matveev

(ramenm, nyashpy)

Python Developer | Backend | AI/ML

4+ years of commercial Python/backend delivery with focus on AI/LLM integrations, RAG pipelines, and tooling.

About me

Python backend developer focused on automation, data-intensive workflows, and AI tooling. I build FastAPI services and integrations that reduce manual work, speed up delivery, and stay maintainable in production.

 Experience (5 years 6 months)

Python Developer (Junior)

Freelance

Junior · Contract

Sep 2019 — May 2021

(1 year 8 months)

Implemented development tasks: from creating small parsers to automating routine processes and writing utility scripts.

Stack: Python, Automation, Web scraping, Data processing

  • Delivered end-to-end tasks from discovery to production-ready tools.
  • Automated repetitive workflows to reduce manual operations and cycle time.
  • Maintained parsers and utility scripts so they stayed reliable after site changes.

R&D Data Collection Engineer

Brightdata

Middle · Full-time

May 2021 — Feb 2023

(1 year 9 months)

Created and maintained tools for advanced web scraping, analysing requests and transforming them for further processing. Injected changes into JavaScript logic for fast and accurate data extraction.

Stack: Python, JavaScript, HTTP, Reverse engineering, Data extraction

  • Built and maintained tooling for advanced, high-friction scraping use cases.
  • Injected JavaScript logic changes to improve extraction speed and reliability.
  • Improved pipeline resilience against anti-bot changes and page-structure drift.

R&D Engineer (Python / AI)

Independent Projects / Research

Middle · R&D

Mar 2023 — Jan 2024

(10 months)

Focused on independent projects and research: Python automation, AI/LLM tooling prototypes, and backend/infrastructure skill development.

Stack: Python, LLM, RAG, FastAPI, Tooling

  • Ran focused R&D tracks in automation and AI/LLM tooling.
  • Built working prototypes that later informed production-grade implementations.
  • Strengthened backend and infrastructure skills for scalable integrations.

Python Developer

Freelance

Senior · Contract

Feb 2024 — present

(2 years 1 month)

Reverse‑engineered websites, built bots and integration solutions to optimize processes, and worked extensively on neural-network/AI/LLM tasks (prototyping, automation, RAG/tooling, model API integrations).

Stack: Python, FastAPI, LLM APIs, RAG, Integrations, Reverse engineering

  • Deliver backend integrations aligned with real business workflows.
  • Ship AI/LLM integrations: prototypes, automation pipelines, RAG, and tool-calling.
  • Reduce time-to-delivery through reusable modules and integration standards.

Skills

Backend

  • Python / asyncio
  • FastAPI / Pydantic
  • PostgreSQL
  • Docker
  • OpenTelemetry

Data / AI

  • Pandas
  • NumPy
  • PyTorch
  • RAG
  • Retrieval / Reranking
  • Vector databases
  • vLLM

Integrations

  • MCP
  • API Analysis
  • Performance

 Projects

Trading bot for an in-game marketplace

Backend / Automation Engineer · 2024 · Solo

Stack: Python, FastAPI, Reverse engineering, Trading automation

  • Built a full loop: API layer, logging, and strategy controls.
  • Kept execution stable under strict platform constraints.

Led the full development cycle: performed reverse engineering of existing solutions, designed the API from scratch, and implemented an automated trading strategy.

Result: Delivered a production-ready automation workflow with stable trade execution, logging, and strategy controls under platform constraints.

Plans: Improve risk controls, monitoring, and resilience; add adapters for new marketplaces behind a unified integration layer.

Universal LLM Models Hub (nnzen.com)

Founder / Backend Engineer · 2025 · Solo

Stack: Python, FastAPI, OpenRouter API, Data aggregation, Frontend table UI

  • Aggregates 300+ model records in production.
  • Auto-enriches model data with leaderboard positions.

Built and launched a FastAPI web platform that consolidates data on 300+ language models from the OpenRouter API — names, context-window length, and current token pricing — and automatically augments them with their positions from the LMArena leaderboard. The result is presented in an interactive searchable table.

Result: The platform is in production: aggregates 300+ open-source models, helping developers quickly find and compare LLMs.

Plans: Automatic tracking of new leaderboard positions and a public API.

Live

Custom agent (LLM core + MCP)

LLM Platform Engineer · 2025 · Solo

Stack: Python, FastAPI, MCP, Tool calling, CLI

  • Implemented plugin hot-reload without restarting the core.
  • Built cascading tool calls with context handover.

Developed a chat assistant for developers: an LLM core and FastAPI server on MCP that hot-loads plugins and manages them, supporting toolchains with context passing.

Result: Delivered a ready-to-use EnhancedMCP core with plugin hot-reload, a CLI client with smart prompter and cascading calls; basic plugin set: file system, multi-LLM, browser automation, terminal, and execution trace history.

Plans: UI development; I/O optimisation; improved prompt engineering (dynamic templates, token optimisation); extended auto-executability (plan-and-execute cycles).

 Contacts

Contact me in any way that works for you:

Work format: remote/hybrid. Open to full-time and contract roles.