Dmitri M.

AI Engineer | Flink | Kafka | Python | Java | AWS

Your AI is only as good as the data behind it. I build both. Most AI projects stall because the data isn't ready: it's messy, slow or stuck in silos. I'm an AI Engineer with a strong Data Engineering background. I build LLM apps and AI agents on top of real-time streaming pipelines and cloud data platforms, so your AI features work in production and don't stop at the demo. Check my skills 👇 🚀 AI Engineering: ✅ LLM application development with OpenAI, Anthropic Claude, and open-source models ✅ RAG (Retrieval-Augmented Generation) pipelines: chunking, embeddings, hybrid search and re-ranking ✅ AI agents and tool-calling workflows using LangChain, LlamaIndex and MCP (Model Context Protocol) ✅ Vector databases: pgvector, Pinecone, Qdrant, OpenSearch and Elasticsearch ✅ Prompt engineering, LLM evaluation, cost and latency tuning ✅ Deploying AI services with FastAPI, Docker and AWS Lambda/ECS 🚀 Streaming & Real-Time Data: ✅ Apache Flink (PyFlink and Java) for stateful stream processing at scale ✅ Apache Kafka for event-driven architectures and real-time ingestion ✅ Real-time feature and embedding pipelines that feed AI and ML systems 🚀 Programming Languages: ✅ Python, Java, Golang and Node.js ✅ Complex SQL, query tuning and database performance optimization 🚀 Cloud Data Platforms: ✅ AWS: Lambda, Step Functions, S3, Glue, Athena, EMR, Redshift, EC2, ECS, ECR and CloudFormation ✅ Lakehouse: Apache Iceberg on S3 ✅ GCP: BigQuery and Looker Studio 🚀 Databases: ✅ SQL: PostgreSQL, MySQL, MSSQL and SQLite ✅ NoSQL: MongoDB, Redis, Elasticsearch and DynamoDB 🚀 Data Stack & DevOps: ✅ Spark, dbt and Apache Airflow for batch processing, data modeling and orchestration ✅ Docker and Kubernetes for containerized deployments ✅ Git, GitHub and GitLab CI/CD 🚀 Data Collection & Automation: ✅ Web scraping with Scrapy, BeautifulSoup, Selenium and Playwright to collect AI training and RAG data ✅ Social media data pipelines for analytics and AI enrichment ⚡ What I can build for you: ✅ AI chatbots and assistants trained on your company's documents and databases ✅ Real-time streaming pipelines that feed AI models and dashboards ✅ Moves from legacy ETL to modern cloud data platforms ✅ End-to-end AI products, from data ingestion to deployed API ⚡ Why clients work with me: ✅ Production-first: I build systems that are monitored, scalable and cost-aware ✅ Clear communication, with regular progress updates and documented code ✅ Flexible hours that overlap with US, EU and APAC time zones 💬 Have an AI idea or a data pipeline problem? Send me a message with a short description of your project. I'll reply with a clear plan, the tech I'd recommend and a realistic timeline, free of charge. Let's turn your data into something your business can use.