Hire the Best CUDA Developers

🏆Top Rated for more than 10 years on Upwork | 🏆 70+ projects | 🏆 100% Job Success Score | 🏆 16 years experience | 🏆Working with 3x enterprise companies and so many start-ups/individual clients Production-grade computer vision systems for real-world environments — edge deployment, system integration, and measurable business outcomes. I graduated as an Electrical Engineer M.Sc. (Embedded Software Developer faculty). I am an expert in AI/Machine learning (object detection) and IoT/Edge Computing (Raspberry Pi and Jetson Orin NX/Nano, Deepstream 6.x, 7.x, Triton/TAO toolkit). I prefer long-term projects, but you can hire me for shorter ones also as a consultant. Strength: - Computer Vision/Machine learning Project architect/design - technical stack, cost of long-term usage, refine user requirement etc. - Python Machine learning projects (Tensorflow, Keras, PyTorch) - especially with computer vision, object detection (YOLO), image segmentation and pose tracking (OpenPose), shape recognition (dlib) and action recognition (X3D, mmaction etc.) - Develop/maintain/upgrade your existing Deepstream/Triton based application - Embedded image processing solutions (Edge computing): computer vision on Jetson Orin (NX/Nano) with Deepstream7.0 (earlier version too 6.x/7.x, Triton, TAO toolkit) or on Raspberry Pi, using Intel Movidius Neural Compute Stick (NCS) + OpenVINO. Train a model for specific purpose, provident dataset collection/preparation/preprocessing task management, I specialized for object detection (up to YOLOv12) - AWS, GCP (Google Cloud Platform) VM inference and training setup and execution/deployment - NVIDIA GPU experience: Tesla T4, L4, A100, H100 GPU for training and inference/server production. - Home automation/Smart home - lighting, building engineering, heating/cooling, Home assistant: integrate Amazon Alexa, Phillips Hue, Sonos, Apple TV, Google Nest, custom solutions - Raspberry Pi (Linux - Raspbian) - programming and teaching in Python (RS232, RS485, TCP communication), Wi-Fi settings, Home assistant Services: - Short (30-60 minutes) consultation - AI Architect tasks: responsible for the architecture of AI solutions, which includes planning, implementing, and managing AI technologies within an organization or project - I can speak English and Hungarian (magyar) and a little bit Spanish (castellano, espanol) AI Computer Vision | MLOPS | Edge Computing | AI Automation | AI strategy | AI consultant

I design, optimize, and integrate real-time object detection and tracking pipelines for NVIDIA Jetson, RK3588, and cloud environments — with measurable gains in FPS, latency, and deployment stability. From YOLO training to TensorRT optimization and production integration, I build Computer Vision systems that work reliably on real hardware. If you need more than just a trained model — if you need a working AI system integrated into hardware or software — I deliver complete, production-ready solutions. 🎯 WHAT I DELIVER • End-to-End Deep Learning Pipelines Model architecture → dataset optimization → training → evaluation → deployment • Real-Time Object Detection & Multi-Object Tracking Optimized YOLO pipelines with stable tracking (DeepSORT / ByteTrack / BOT-SORT) • ⚡ Edge AI Acceleration & Performance Optimization TensorRT conversion, CUDA acceleration, latency reduction, memory tuning, FPS improvements • 🔗 AI Integration into Production Systems Jetson deployment, inference APIs, embedded integration, debugging, monitoring & system optimization 🏆 PROVEN EXPERIENCE • 5★ Upwork reviews for NVIDIA Jetson Nano, AGX & Orin deployments • Deployed Frigate and real-time CV pipelines on Jetson Orin NX • Optimized inference performance using TensorRT for improved real-time execution • Installed and configured LLM environments with secure key management & usage tracking • Research & production AI deployment experience at HBrain 🛠 TECH STACK Deep Learning: PyTorch, CNN architectures, YOLO variants Computer Vision: OpenCV, real-time video processing, detection & tracking Optimization: TensorRT, CUDA Systems: Linux, Embedded Systems, NVIDIA Jetson Languages: Python, C++ If you share your hardware target, dataset sample, or performance goal, I can propose a clear technical architecture and realistic delivery plan.

I help startups and AI companies build high-performance AI systems, GPU-accelerated applications, and scalable AI infrastructure using CUDA, Rust, and Python. My expertise focuses on performance engineering, AI infrastructure, parallel computing, and low-latency systems designed for production-scale workloads. I work on optimizing compute-heavy applications, improving inference performance, and building reliable backend systems for modern AI products. I specialize in: • CUDA & GPU Optimization • AI Inference Optimization • Rust-based High Performance Systems • AI Infrastructure Engineering • Parallel Computing • Low-Latency Backend Systems • Python-based AI & Automation Systems • LLM Infrastructure • RAG Architecture & AI Search • Scalable API & System Architecture I can help with: ✅ GPU acceleration and CUDA optimization ✅ AI inference performance tuning ✅ Rust backend systems for high-performance workloads ✅ AI infrastructure and deployment pipelines ✅ Memory and compute optimization ✅ Parallel processing systems ✅ AI search and RAG pipelines ✅ Python automation and backend development ✅ Scalable distributed systems ✅ Production-ready AI architecture Tech Stack & Tools: CUDA, Rust, Python, PyTorch, TensorRT, ONNX, Docker, FastAPI, PostgreSQL, Vector Databases, REST APIs, WebSockets, AWS, Azure, and cloud-native architectures. I focus on building systems that are: * performant * scalable * efficient * reliable * production-ready If you are building GPU-intensive applications, AI infrastructure, inference systems, or performance-critical software, I’d be happy to discuss your project. Send me a message, and let’s discuss how we can make it happen!

Most AI developers can build a prototype. Very few can optimize and deploy multimodal LLMs on Jetson Orin Nano at real-time speeds, architect enterprise-grade RAG systems across 3,000+ SQL tables, fine-tune state-of-the-art vision-language models, or serve high-throughput inference with vLLM and SGLang at production scale. That’s what I do. I’m a Senior AI/ML Engineer specializing in: • Production LLM Systems & Agentic AI • Edge AI & High-Performance Model Optimization • Computer Vision & Real-Time Multimodal AI • Distributed Inference & AI Infrastructure I don’t just wrap APIs I engineer AI systems from the model architecture and optimization layer all the way to scalable production deployment. What makes my work different: → I build multi-agent AI systems that feel invisible to users. Designed a dual-LLM tutoring architecture where a speech-to-speech AI tutor interacts live with students while a hidden orchestration LLM performs real-time prompt injection, memory routing, retrieval planning, and response control completely transparently. → I optimize and deploy AI where most teams fail. Fine-tuned PaliGemma, converted it to ONNX, applied INT8/FP16 quantization and TensorRT optimization, reducing model size by 65% with <2% accuracy loss, then deployed it on Jetson Orin Nano, Raspberry Pi, edge devices, and mobile hardware for real-time inference with extremely limited compute. → I build enterprise RAG systems that reason not just retrieve. Architected a multi-stage multilingual RAG pipeline using HyDE, recursive retrieval, schema-aware chunking, metadata filtering, hybrid search, reranking, graph-based retrieval, chain-of-table reasoning, and long-context orchestration across 3,000+ enterprise SQL tables and 150+ warehouses. → I engineer high-performance LLM inference infrastructure. Built scalable inference pipelines using vLLM, SGLang, ONNX Runtime, TensorRT-LLM, DeepSpeed, FlashAttention, speculative decoding, paged attention, KV-cache optimization, and distributed GPU serving for low-latency, high-concurrency AI systems. → I deliver video AI and lip sync systems beyond standard open-source baselines. Fine-tuned LatentSync on a 4,000-sample dataset, outperforming Wav2Lip, GAN-Wav2Lip, and Wav2LipHD baselines in multilingual real-time video translation with highly coherent visual speech synchronization. → I design AI systems for real-world production environments. Built AI backends with FastAPI, asyncio, WebSockets, Redis queues, Kafka streaming, GPU worker orchestration, autoscaling inference services, and Kubernetes-based deployments capable of handling real-time concurrent workloads. What I build for clients: • Production-grade RAG systems with hybrid, recursive, graph, and agentic retrieval • High-throughput LLM serving using vLLM, SGLang, TensorRT-LLM, Triton Inference Server • Edge AI deployment with ONNX, TensorRT, OpenVINO, CoreML, TensorFlow Lite, ONNX Runtime • Fine-tuning pipelines (LoRA, QLoRA, PEFT, RLHF, DPO) with evaluation and benchmarking • AI agents with LangGraph, LangChain, LlamaIndex, CrewAI, AutoGen, MCP architectures • Multimodal AI systems combining vision, audio, speech, and language models • Text-to-SQL systems for complex enterprise schemas with multilingual querying • Computer Vision pipelines YOLOv8, YOLO11, SAM2, GroundingDINO, DETR, Mask R-CNN • Real-time speech AI Whisper, XTTS, RVC, voice cloning, speech-to-speech pipelines • Video AI lip sync, face reenactment, multilingual dubbing, avatar systems • Distributed AI infrastructure with Docker, Kubernetes, Ray, Celery, Redis, Kafka • AI observability, evaluation, tracing, and monitoring with LangSmith, MLflow, Weights & Biases • Vector databases and retrieval systems FAISS, Pinecone, ChromaDB, Milvus, Qdrant, Weaviate • AI surveillance and geospatial intelligence systems with real-time threat detection Core Stack: Python · PyTorch · TensorFlow ·  HuggingFace · Transformers · LangChain · LangGraph · LlamaIndex · OpenAI · Anthropic · Gemini · FastAPI · vLLM · SGLang · TensorRT-LLM · Triton · ONNX Runtime · TensorRT · OpenVINO · TensorFlow Lite · CUDA · OpenCV · MediaPipe · Whisper · YOLO · SAM · GroundingDINO · FAISS · Pinecone ·  Qdrant · MongoDB · PostgreSQL · Redis · Kafka ·  Docker · Kubernetes · GCP · Vertex AI · AWS · Azure · WebSockets If you need an AI engineer who can go deeper than most teams whether that means deploying optimized multimodal AI on constrained edge hardware, architecting enterprise RAG systems that minimize hallucinations, building scalable LLM inference infrastructure, or shipping production-grade multi-agent AI send me a message. I’ll tell you honestly what’s possible, what’s overhyped, and exactly how I’d build it.

Many computer vision models work in a notebook but fail in production. I build OCR, detection, and tracking systems that actually run in real environments so your team can automate workflows and extract real business value from visual data. I work with companies and startups who need robust AI pipelines for document automation, traffic monitoring, retail analytics, or custom visual AI applications. If you want a low-budget experiment with no clear success criteria, I’m probably not the best fit. 🔎 𝗛𝗼𝘄 𝗜 𝗪𝗼𝗿𝗸 𝟭. 𝗜𝗻-𝗗𝗲𝗽𝘁𝗵 𝗗𝗶𝘀𝗰𝗼𝘃𝗲𝗿𝘆 I begin each project by understanding your goals, constraints, and success metrics to ensure the solution meets your unique needs. 𝟮. 𝗠𝗼𝗱𝘂𝗹𝗮𝗿 𝗣𝗿𝗼𝗯𝗹𝗲𝗺-𝗦𝗼𝗹𝘃𝗶𝗻𝗴 I break complex tasks into smaller parts and test multiple approaches to find the most effective solution. This ensures reliable results you can trust. 𝟯. 𝗟𝗲𝘃𝗲𝗿𝗮𝗴𝗶𝗻𝗴 𝗔𝗱𝘃𝗮𝗻𝗰𝗲𝗱 & 𝗖𝘂𝘀𝘁𝗼𝗺 𝗠𝗲𝘁𝗵𝗼𝗱𝘀 My toolkit includes all major computer vision tasks: Classification, detection, tracking, and segmentation. I combine off-the-shelf models with custom-built methods to achieve top-notch performance. 𝟰. 𝗖𝗹𝗲𝗮𝗿, 𝗙𝗿𝗲𝗾𝘂𝗲𝗻𝘁 𝗖𝗼𝗺𝗺𝘂𝗻𝗶𝗰𝗮𝘁𝗶𝗼𝗻 I keep you updated at every step, provide realistic timelines, and immediately address any hurdles. Even if you’re not technical, I’ll explain everything in plain language so you always know where your project stands. 𝟱. 𝗔𝗹𝘄𝗮𝘆𝘀 𝗣𝘂𝘁𝘁𝗶𝗻𝗴 𝗬𝗼𝘂 𝗙𝗶𝗿𝘀𝘁 My clients consistently give me 5-star ratings and glowing feedback. If your requirements stretch beyond my skill set, I’ll be transparent and let you know right away. 𝗔𝗱𝘃𝗮𝗻𝗰𝗲𝗱 𝗖𝗼𝗺𝗽𝘂𝘁𝗲𝗿 𝗩𝗶𝘀𝗶𝗼𝗻 & 𝟯𝗗 𝗣𝗲𝗿𝗰𝗲𝗽𝘁𝗶𝗼𝗻 For projects involving robotics, autonomous systems, or spatial analytics, I also build 3D perception pipelines using LiDAR, stereo cameras, and point clouds. This includes 3D object detection, Bird’s Eye View (BEV) transformations, and point cloud processing using deep learning. 𝗥𝗲𝗰𝗲𝗻𝘁 𝗣𝗿𝗼𝗷𝗲𝗰𝘁𝘀 𝟭. 𝗠𝘂𝗹𝘁𝗶-𝗔𝗣𝗜 𝗢𝗖𝗥 𝗣𝗶𝗽𝗲𝗹𝗶𝗻𝗲: Combined Google Vision, OpenAI, and AWS Rekognition to increase document extraction accuracy across noisy images. 𝟮. 𝗦𝗰𝗮𝗻𝗻𝗲𝗱 𝗗𝗼𝗰𝘂𝗺𝗲𝗻𝘁 𝘁𝗼 𝗘𝘅𝗰𝗲𝗹: Parsed key fields using Python + QwenVL and auto-generated Excel reports. 𝟯. 𝗟𝗶𝗰𝗲𝗻𝘀𝗲 𝗣𝗹𝗮𝘁𝗲 𝗗𝗲𝘁𝗲𝗰𝘁𝗶𝗼𝗻: End-to-end system with YOLOv8 trained on custom dataset and deployed for inference 𝟰. 𝗥𝗲𝗮𝗹-𝗧𝗶𝗺𝗲 𝗟𝗶𝗗𝗔𝗥 𝗢𝗯𝗷𝗲𝗰𝘁 𝗗𝗲𝘁𝗲𝗰𝘁𝗶𝗼𝗻 & 𝗧𝗿𝗮𝗰𝗸𝗶𝗻𝗴 Built a 3D detection and tracking pipeline using LiDAR and camera data with 3D bounding boxes, frame-to-frame association, and 2D/3D visualizations for autonomous navigation. 𝟱. 𝗗𝗲𝗲𝗽 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗼𝗻 𝗣𝗼𝗶𝗻𝘁 𝗖𝗹𝗼𝘂𝗱𝘀 Trained PointNet and voxel-based 3D CNNs on ShapeNet Core for point cloud segmentation and classification, including full preprocessing and model visualization. 𝗧𝗲𝗰𝗵 𝗦𝘁𝗮𝗰𝗸 I work with Python, PyTorch, TensorFlow, OpenCV, YOLO, Open3D, and modern vision APIs (Google, AWS, OpenAI) to build detection, tracking, and OCR systems. 𝗪𝗵𝗮𝘁 𝗖𝗹𝗶𝗲𝗻𝘁𝘀 𝗦𝗮𝘆 "Yacine is reliable, very good at his job, and very informative. He was able to set up a POC, identify the main pitfalls, and propose solutions independently." "Yacine is committed to provide high quality work. He knows what he's doing. It's a pleasure to work together. I recommend him for data mining and vision work." "Yacine always does a great job on any computer vision related task, he delivered the project very quickly. I will definitely rehire him again whenever needed." 📬 Let’s Talk Send a message describing your computer vision problem and the data you’re working with. If it’s a good fit, we’ll discuss the next steps.

Experience in: Speech to Text Conversion Text to Speech Synthesis Natural Language Processing (NLP) Computer Vision and Neural Networks CUDA, PyCuda I'm a Machine Learning Engineer passionate about innovation and eager to work in Japan's technology sector. With expertise in various areas such as speech processing and GPU computing, I am looking for opportunities to contribute my skills in the Japanese market. I would also appreciate assistance with the work visa process for potential relocation. Feel free to reach out for collaboration or inquiries.