Bakhtiyor Sulaymonov

ML Engineer | LLMs | NLP | AI integration · Tashkent, Uzbekistan

ML Engineer specializing in NLP and production ML deployment. I combine 6+ years of software engineering experience with hands-on expertise in fine-tuning transformer models, building RAG systems, and creating automated data pipelines.

Transitioned from software development to AI/ML engineering, bringing strong software expertise to machine learning applications. I focus on deploying robust, scalable solutions that bridge the gap between research and production.

Passionate about continuous learning, clean code, and solving real-world problems with AI.

bakhti.ai.uz@gmail.com · LinkedIn · GitHub

Skills

Languages: Python, SQL, JavaScript

ML/AI: PyTorch, CUDA, Hugging Face (Transformers, Datasets), Pandas, NumPy, Azure OpenAI, RAG, Vector DBs, Prompt Engineering, NLP, LLMs

Tools & Frameworks: FastAPI, Selenium, Git, Linux, Bash, pytest, uv, ruff, Vue, React, Nuxt.js

Recent Projects (AI/ML)

Production NLP Text Classification System

Fine-tuned and deployed a RoBERTa-based text classification model for production use at a government agency. Conducted comparative analysis of transformer architectures, generated synthetic training data using LLMs, and built a complete document processing pipeline.

PyTorch, CUDA, Hugging Face, FastAPI

RAG-Based AI Chatbot System

Deployed a Retrieval-Augmented Generation chatbot integrated with official government Telegram bot using Azure OpenAI. Designed vector database architecture, engineered system prompts with guardrails, and built monitoring infrastructure.

Azure OpenAI, Vector Database, Telegram Bot API, Python

Automated Data Collection Pipeline

Built automated web scraping system for a conversational dataset collection from social media platforms. Implemented robust error handling, automated as weekly cron job, and built data preprocessing pipeline.

Python, Selenium, Bash, Linux, Chrome

Parallel Translation Dataset

Created parallel dataset for machine translation tasks by processing and aligning multilingual data from public sources. Implemented alignment algorithms, data cleaning, and quality filtering pipelines.

Python, Pandas, Hugging Face Datasets, NumPy

Experience

AI/ML Software Engineer

Ministry of Justice, Tashkent

January 2025 – January 2026

Software Engineer

Vention, Remote

February 2022 – February 2024

Software Engineer / Team Leader

DataPrizma, Tashkent

February 2019 – January 2022

Frontend Developer

SAB Games, Tashkent

October 2018 – February 2019

Education

Politecnico di Torino, Italy

B.Sc. Computer Engineering (2013 – 2018)

Focus area on Computer architecture, Databases, Networking, OOP, Algorithms