My core specialization lies in building complex systems centered around Large Language Models (LLMs). With deep expertise in Python backend development, machine learning, DevOps, and data engineering, I create end-to-end solutions that I can deliver independently. However, I also collaborate with a network of professionals in related fields who can assist or accelerate development (e.g., frontend development, UX/UI, security configurations).
Skills:
I excel in Python programming and have extensive experience with frameworks and technologies such as FastAPI and Flask, enabling the creation of applications that fully adhere to RESTful API standards, as well as solutions based on WebSockets. My skills also extend to database management with MongoDB and PostgreSQL. I employ modern approaches for validating incoming data using libraries like Pydantic, and I utilize Redis for caching. I am proficient in creating solutions with delayed task execution using Celery, including ensuring tasks are processed asynchronously. Recently, I have primarily focused on developing asynchronous applications, though I understand when to use asynchronous vs. synchronous approaches, and when to opt for multiprocess or multithreaded solutions. To ensure fast, efficient deployment of scalable applications, I leverage Docker. For solutions requiring high reliability in data processing and guaranteed message delivery, I use message brokers like RabbitMQ or Kafka.
My machine learning specialization focuses on Large Language Models (LLMs). I have a proven track record of developing both simple and complex bots using advanced tools such as LangChain, LangGraph, and LlamaIndex. I have successfully implemented Retrieval-Augmented Generation (RAG) systems that intelligently respond based on knowledge base data, improving the accuracy and relevance of automated responses. I have developed systems that extract structured data from unstructured sources (pdf, xls, docs, txt) for efficient information retrieval using semantic search via vectors or graphs. Additionally, I have created advanced conversational agents using agent-based frameworks, significantly expanding the capabilities of these systems in dynamic, real-world scenarios.
I can build solutions on your servers or in the cloud.
My data engineering skills are demonstrated through the use of AWS services, including Glue, Athena, Aurora, RedShift, and Quicksight, as well as data orchestration with Airflow and data integrity management with dbt and Great Expectations. Advanced analytics and big data processing are facilitated through platforms like Spark, Snowflake, and integrations such as AWS Lambda Functions and OpenSearch.
Recent Achievements:
- Under my leadership, the technical team developed an updated core product for the platform, which accelerated client bot development by 40 times and reduced the cost of such bots by 80 times.
- I took the initiative to develop a transparent and fair system for evaluating our technical specialists, which also includes an effective method for calculating salaries. This system ensures that everyone's contributions are recognized and valued, fostering a positive work environment.
- Thanks to the development standards I established, the deployment of updates has accelerated 10 times.
Technical skills: Python, FastAPI, Flask, RESTful API, MongoDB, PostgreSQL, Mongo Engine, Pydantic, Celery, Docker, Alembic, Redis, RabbitMQ, Unittests, pytest, OpenTelemetry, LLM, LangChain, LangGraph, LlamaIndex, ElasticSearch, Git, GitHub Actions, Streamlit, Nginx, CI/CD with GitLab, AWS Glue, AWS Athena, AWS Aurora, AWS StepFunctions, AWS RedShift, AWS Quicksight, AWS S3, AWS Lambda Functions, OpenSearch, Spark, Snowflake, dbt, Great Expectations, numpy, Airflow, React, Full Stack.
