Tsimafei S.

𝗔𝗰𝗰𝗼𝗺𝗽𝗹𝗶𝘀𝗵𝗲𝗱: As a RAG Developer, built multi-agent RAG with SQL access and CRM integration via LangChain and LlamaIndex - replaced 3 support staff, response time from hours to seconds. Set up vector DBs: Pinecone, Weaviate, Milvus. Integrated Whisper and AssemblyAI for call transcription and diarization. Built AI scraping pipelines with Playwright and LLM parsing. rag developer, llm developer, langchain developer, llamaindex developer, langgraph developer, crewai developer, pinecone developer, weaviate developer, milvus developer, pgvector developer, prompt engineering, openai whisper developer, assemblyai developer, speaker diarization, call transcription, playwright developer, selenium developer, ai scraping engineer, ai engineer saas, enterprise rag system, n8n developer, make com automation

I'm an AI Engineer leading ShiftLine - 90+ specialists delivering RAG systems, LLM integrations, computer vision, and MLOps for SaaS, fintech, and manufacturing clients across the US and Europe. As an AI Engineer and ML Engineer, I oversee the AI/ML division and 17-engineer DevOps team. As an Agentic AI Developer, I build autonomous workflows using LangGraph, CrewAI, and AutoGen. As an LLM Engineer, I handle fine-tuning and self-hosted deployment via LoRA, QLoRA, and vLLM. ai engineer, rag developer, llm developer, agentic ai developer, ml engineer, mlops engineer, computer vision engineer, nlp engineer, langchain developer, langgraph developer, crewai developer, llamaindex developer, llm fine-tuning, lora qlora, vllm engineer, rag pipeline, hugging face, fastapi developer, n8n developer, make com automation, kubernetes engineer, aws engineer, pytorch developer, tensorflow developer

𝗣𝗿𝗼𝗷𝗲𝗰𝘁: DevOps transformation for Leroy Merlin. 𝗔𝗰𝗰𝗼𝗺𝗽𝗹𝗶𝘀𝗵𝗲𝗱: As a DevOps Engineer and MLOps Engineer, took deployment from 1/month to 120/month, cut deployment time 85%, reduced infrastructure costs 55%. Full Kubernetes + CI/CD + monitoring stack on AWS, GCP, Azure. Deploy production AI via Docker and FastAPI. Manage self-hosted LLM inference with vLLM and DeepSpeed. devops engineer, mlops engineer, kubernetes engineer, ai infrastructure engineer, aws engineer, google cloud engineer, azure engineer, terraform developer, ansible engineer, jenkins engineer, docker engineer, fastapi engineer, kafka engineer, grafana engineer, prometheus engineer, ci/cd engineer, sre engineer, finops engineer, aiops engineer, vllm engineer, deepspeed engineer, llm infrastructure, self-hosted llm, cloud cost optimization

I'm an LLM Engineer and I worked on early LLM integration projects before joining PlaysDev full-time. As an AI Automation Developer, I built GPT-based workflows for content generation, data extraction, and customer support automation. As an LLM Engineer, I handled prompt engineering, API integration, and cost optimization for OpenAI-based systems using n8n and Make com. llm engineer, ai automation developer, gpt developer, openai developer, prompt engineering, n8n developer, make com automation, ai workflow automation, langchain developer, content generation ai, customer support ai, llm integration, api integration engineer

I'm an ML Engineer and Computer Vision Engineer delivering production AI systems for fintech, insurance, and manufacturing. Deployed YOLOv8, YOLOv10, Detectron2 for real-time QC on production lines — defect escape rate near zero. Built NLP pipelines using BERT, RoBERTa, SpaCy. Automated document processing with AWS Textract and LayoutLM. Built churn prediction and forecasting models with XGBoost and CatBoost. computer vision engineer, ml engineer, ai engineer, yolov8 developer, yolov10 developer, detectron2 developer, opencv developer, bert developer, roberta developer, spacy developer, hugging face developer, aws textract developer, layoutlm developer, xgboost developer, catboost developer, scikit-learn developer, predictive analytics, churn prediction, document processing ai, nlp engineer, pytorch developer, tensorflow developer

𝗣𝗿𝗼𝗷𝗲𝗰𝘁: Cloud migration and infrastructure modernization for a European fintech company. 𝗔𝗰𝗰𝗼𝗺𝗽𝗹𝗶𝘀𝗵𝗲𝗱: Migrated legacy on-premise infrastructure to AWS — reduced infrastructure costs by 40% and improved uptime from 97% to 99.9%. Built full Kubernetes cluster with auto-scaling, CI/CD pipelines via Jenkins, and IaC using Terraform and Ansible. devops engineer, kubernetes engineer, cloud migration engineer, aws engineer, terraform developer, ansible engineer, jenkins engineer, docker engineer, ci/cd engineer, infrastructure as code, cloud cost optimization, fintech devops, platform engineer, site reliability engineer, sre engineer

𝗣𝗿𝗼𝗷𝗲𝗰𝘁: MLOps infrastructure buildout for a US-based ML team of 15 engineers. 𝗔𝗰𝗰𝗼𝗺𝗽𝗹𝗶𝘀𝗵𝗲𝗱: Built end-to-end MLOps pipeline — experiment tracking, model versioning, automated retraining, and production deployment via FastAPI and Docker on GCP. Reduced model deployment time from 2 weeks to 1 day. Set up full observability stack with Grafana and Prometheus for real-time model performance monitoring. mlops engineer, ai infrastructure engineer, model deployment engineer, fastapi engineer, docker engineer, gcp engineer, grafana engineer, prometheus engineer, experiment tracking, model versioning, automated retraining, ml pipeline engineer, kubernetes engineer, ci/cd engineer, devops engineer

I'm an ML Engineer and I led R&D projects in computer vision and predictive analytics for manufacturing and retail. As a Computer Vision Engineer, I prototyped object detection systems using OpenCV and YOLO. As a Predictive Analytics Engineer, I built forecasting and scoring models using XGBoost and Scikit-learn deployed via FastAPI on AWS. ml engineer, computer vision engineer, yolo developer, opencv developer, xgboost developer, scikit-learn developer, predictive analytics, fastapi developer, aws engineer, pytorch developer, r&d engineer, ai engineer manufacturing