Hire the Best Deep Learning Experts
in Georgia

An expert in AI, Machine Learning, Data Science, Deep Learning, Computer vision, Image processing, algorithm development and deployment with 2+ years of hands-on experience. I am a freelancer and a graduate researcher in Artificial Intelligence applications. Programming Languages: ✔ Python ✔ Javascript Data Science and web crawling: ✔ NumPy ✔ Scrapy ✔ BeautifulSoup ✔ Selenium ✔ Pandas ✔ SciPy Visualization: ✔ Matplotlib ✔ Seaborn ✔ Plotly Machine Learning: ✔ Pytorch ✔ Scikit-Learn ✔ Keras Computer Vision: ✔ Pillow ✔ OpenCV ✔ Scikit-Image ✔ Tesseract ✔ Dlib Deployment: ✔ Flask(Api, Rest Api) Front: ✔ HTML5 ✔ CSS ✔ Bootstrap ✔ React Neural Networks: ✔ GAN ✔ CNN ✔ ANN ✔ Yolo ✔ FastRCNN ✔ MaskRCNN ✔ Encoder Databases: ✔ Postgres ✔ MySQL ✔ Mongodb ✔ SQLite ✔ PL/SQL ✔ Faiss Languages: ✔ Russian ✔ Georgian ✔ English Soft Skills: ✔ Teamwork ✔ Leadership ✔ Self-motivation ✔ Communication Have already written pytorch implementation of Triplet Loss, Xgan - google GAN network for generating avatars. Also done projects in face similarity, and objects detection and localization. I also have an experience in data mining. I have developed a wide range of software solutions to crawl and extract different types of data from various websites, e.g. real-time data, business listings, sports results data and other related data sources. You can see my projects below:

Hello, I am investing all my time and resources in Upwork ☝ My experience covers data analysis, AI/ML model training, fine-tuning, and deployment to production on AWS, GCP, Azure, or edge devices. ⬣ Skills : GenAI : RAG, Vector databases, LLM finetune, AI Agent/Multi Agent systems. Machine Learning : classification, regression, similarity search. Computer vision : object detection&tracking, pose estimation, image processing. ⬣Programming languages : Python, MATLAB,C#. ⬣ ML/DL LIBRARIES : TensorFlow, Scikit-Learn, Keras, Pandas, Numpy, OpenCV,Pytorch, HuggingFace,Unsloth, Ultralytics. ⬣ Inference engines: llama.cpp, OLlama, LiteRT-LM, TensorRT. ⬣ Certificates : ✅ AWS Certified Solutions Architect Professional ✅DeepLearning.AI Machine Learning Engineer for production I AM READY TO IMPLEMENT YOUR PROJECT AND CONVERT YOUR IDEAS INTO A REALITY!

😱 𝗟𝗲𝘁’𝘀 𝗷𝘂𝗺𝗽 𝗼𝗻 𝗮 𝑭𝑹𝑬𝑬 𝑪𝑨𝑳𝑳 𝗮𝗻𝗱 𝗱𝗶𝘀𝗰𝘂𝘀𝘀! 😱😱😱😱😱😱😱😱 so you can see my approach before deciding. I build production-ready AI and trading automation systems for clients who need more than a simple script. I am an AI/ML engineer with a Master’s degree in Artificial Intelligence and hands-on experience building trading automation, prediction-market tools, real-time data pipelines, dashboards, and AI-powered analysis systems. I help traders, founders, and technical teams turn trading ideas into working software: from data collection and backtesting to live monitoring, bot execution, alerts, dashboards, and AI-based decision support. My work combines AI/ML engineering, real-time market-data infrastructure, trading bots, dashboards, and deployment. I can help you move from an idea or manual workflow to a working system that collects data, analyzes signals, executes logic, sends alerts, stores results, and runs reliably on a VPS or cloud server. Recent project areas include Kalshi and Polymarket prediction-market systems, Webull bot interfaces, real-time WebSocket feeds, Telegram/Discord trading controls, ClickHouse/Grafana dashboards, Docker deployments, and AI-powered market-analysis pipelines. My recent work includes building and improving systems for Kalshi, Polymarket, Webull, and crypto/prediction-market workflows, including real-time WebSocket data processing, order book monitoring, Telegram/Discord bot interfaces, market-data dashboards, database pipelines, and Docker-based deployment on VPS/cloud infrastructure. What I can help you build Trading bots and automation tools Kalshi, Polymarket, Webull, crypto, and broker/API integrations Real-time market-data pipelines WebSocket order book, ticker, trade, and price-feed processing Backtesting and strategy research tools Telegram/Discord trading bots and alert systems AI-powered market analysis and signal-generation workflows Dashboards for monitoring trades, PnL, token usage, prices, and system health Dockerized deployment, logging, debugging, and VPS/cloud setup Data storage using PostgreSQL, ClickHouse, and related tools AI/ML experience I also have strong experience in deep learning and applied AI, including: PyTorch, TensorFlow, Keras, scikit-learn LLM and AI-analysis pipelines generative models: Audio/music/speech AI systems Computer vision and object detection Data analysis, feature engineering, and model evaluation Python automation and production ML workflows Trading-system experience I have worked on systems involving: Kalshi prediction-market bots Polymarket data monitoring and dashboards Webull bot interfaces Real-time price feeds I focus on practical, production-oriented development: clean Python code, clear communication, reliable deployment, and systems that can be monitored, debugged, and improved over time. I can work on trading infrastructure involving real-time market data, WebSocket feeds, order books, execution workflows, monitoring dashboards, and low-latency automation. For more advanced trading architecture such as HFT-style systems, OMS/EMS design, and institutional execution workflows. If you have a trading idea, AI workflow, automation task, or market-data problem, I can help you turn it into a working system. # Technical stack Python, PyTorch, TensorFlow, scikit-learn, Pandas, NumPy, PostgreSQL, ClickHouse, Grafana, Docker, Linux VPS, WebSockets, REST APIs, Telegram bots, Discord bots, OpenAI API, trading APIs, and AI/ML pipelines. I work best with clients who want clear communication, fast iteration, and practical engineering. I can help with research prototypes, MVPs, production grade bots, dashboards, infrastructure, or debugging existing systems. # Keywords : Algorithmic Trading, Algo Trading, Quant Trading, Quant Developer, Trading Bot, Trading Automation, Trading Strategy, Strategy Development, Backtesting, Live Trading, Paper Trading, HFT, Low Latency, Market Making, Order Book, Level2 Data, Level3 Data, Tick Data, Market Data, Execution Engine, OMS, EMS, DMA, Smart Order Routing, Risk Management, Position Sizing, Portfolio Management, Portfolio Optimization, Signal Generation, Alpha Research, Statistical Arbitrage, Pairs Trading, Mean Reversion, Momentum Trading, Trend Following, Scalping, Arbitrage, Options Trading, Futures Trading, Forex Trading, Crypto Trading, Equities Trading, Prediction Markets, Kalshi, Polymarket, Webull, Interactive Brokers, IBKR, Alpaca, Binance, Coinbase, MetaTrader, MT4, MT5, MQL4, MQL5, TradingView, Pine Script, NinjaTrader, QuantConnect, Lean, NautilusTrader, cTrader, FIX API, Broker API, Exchange API, REST API, WebSocket, Trade Execution, Order Routing, Order Management, Fill Handling, Slippage, Latency Optimization, Market Microstructure, Volatility, Greeks, Delta Hedging, Options Spreads, Credit Spreads, Debit Spreads, Short Premium, Risk Controls, Stop Loss, Take Profit, Trailing Stop, PnL Tracking, Trade Alerts, Telegram Bot, luxalgo

I am a passionate mobile developer working both in native and cross-platform (Flutter). I have 5+ years of enterprise experience and a double Master's degree in Business Information Systems. My credo: • Clean code; • Beautiful apps; • Transparent working process My core skills are: − Client-server applications; − Machine learning; − Video streaming; − Augmented reality (AR).

Machine Learning and AI Engineer. 8+ years shipping applied AI into production: LLM agents and RAG, NLP, speech, computer vision, and CUDA. I build systems that hold up in production, not demos, and every claim below is a number I measured. LLM and NLP - Semantic CV-to-job matching for an EU job board: 10,001 CVs against 10,003 vacancies, 100M cosine pairs in about 18s on one RTX 4090 (bge-m3), then a two-stage LLM re-rank that spreads the top 10 from 0.93 down to 0.71 where the raw cosine moves only 0.03. - LLM CV builder: messy PDF / DOCX / LinkedIn text into a strict Pydantic record bound to the client's 674-value taxonomy, so invalid output is structurally impossible. Emits a recruiter-ready LaTeX PDF and database-ready JSON. - Crypto signals from tweet embeddings (live client program): Qwen3-Embedding-8B on an A40, 1M+ tweet vectors in Qdrant, 44 symbols scored every 60 seconds, 4.65M events ingested, 10-day zero-error run. Speech - On-prem call-centre speech-to-text, in production 2020-2022, feeding fraud detection and call routing. Benchmarked on hand-labeled calls: Whisper 20% median WER, Vosk 28%, Google 32%. - Diarized transcription toolkit, TTS, and cloned-voice agents. Computer vision and CUDA - Multi-camera 3D capture for film relighting: 33 datasets delivered, 3-32 synchronized cameras each, GPU space carving at 1.98 s/frame versus 65 s in vectorized NumPy (about 33x). 5.0 Upwork review. - GPU brute-force passphrase recovery: about 58,000 candidates/s on one RTX 4090, 160x over CPU, sharded across multiple GPUs and machines. - Open-set retail robot vision on ROS 2; banknote detection at the cash desk. YOLO, Grounding DINO, RealSense. How I work Anti-hallucination is enforced in the code path, not in a prompt: strict schema validation, retry on violation, closed enum vocabularies. Success criteria are pre-registered before results exist, models are judged on time-ordered splits, and negative results get reported rather than buried. Stack - LLM and agents: OpenAI API, Claude Agent SDK + MCP, LangChain / LangGraph, RAG, vLLM, Hugging Face Transformers, fine-tuning, structured outputs with Pydantic + instructor, hand-written tool-use loops, chatbots on Telegram and web - NLP and search: embeddings, semantic search, vector databases (Qdrant), clustering, topic discovery, classification, multilingual pipelines - Speech: Whisper, Vosk / Kaldi, Deepgram, Google Cloud STT, TTS, speaker diarization - Vision: OpenCV, YOLO, Grounding DINO, TensorFlow Object Detection, RealSense RGB-D, ROS 2, NVIDIA Jetson - Classic ML: PyTorch, TensorFlow, scikit-learn, CatBoost, forecasting, time-series validation, reinforcement learning - GPU: CUDA C++, PyCUDA, Nsight Compute, multi-GPU sharding - Data and infra: Python, FastAPI, Docker Compose, PostgreSQL, Redis, BigQuery, pandas / NumPy / Parquet, n8n, Linux VPS, GCP Full case studies for each project above, with architecture diagrams and honest results including what did not work, are in my Portfolio section below. Tell me what you are building and I will tell you plainly whether I am the right fit for it.

I'm a data scientist and AI engineer with a few years of production experience with predictive modeling, multi-agent systems, RAG pipelines, end-to-end deployment. Some things I've shipped: an energy expert chatbot for politicians and policy staffers, a multi-agent customer support system across email, chat and SMS, and a regressive income prediction model approved by the National Bank of Georgia for a 10M GEL pilot. I work in Python primarily, stack includes: LangChain, LangGraph, FastAPI, scikit-learn, XGBoost, Supabase, Pinecone. On the frontend with React/TypeScript. I've also shipped a production iOS app in Swift. I care about building things properly, not just getting them to work once. If you've got a serious project, let's talk.