Cuda jobs

... — but with a closed, gaming-style enclosure and a proper integrated power button. Primary use: running AI image-recognition models locally. I'll handle all AI software myself — this job is hardware only: choosing the parts, building the unit, and delivering it with Windows installed and drivers working. The main thing this affects for you is component choice: an NVIDIA dGPU with CUDA support and as much VRAM as the budget allows is the priority. Secondary goal: noticeably cheaper than an equivalent desktop PC while staying close in performance. Deliverable One fully assembled, tested, working unit shipped to me, plus a parts list, wiring diagram, photos of the internals before closing the case, and a short setup/maintenance guide. Core components Laptop motherbo...

I need someone to jump on AnyDesk or TeamViewer and turn my Windows 11 desktop with an NVIDIA GPU into a fully portable DeepFaceLive box. Start by checking which CUDA toolkit is already present—I honestly don’t know—then install or upgrade whatever version best matches the GPU and DeepFaceLive’s current build. Once CUDA is sorted, set up a clean, self-contained (no system-wide edits) Python environment; I’m flexible on the exact 3.x version as long as every required library plays nicely with the GPU. Finish by enabling the virtual camera so I can push the swapped feed into Zoom, OBS or Discord without additional tweaks. Acceptance test • Launch DeepFaceLive from the portable folder you create • Load a sample model and run a live...

My 3D scanner streams PCD point-cloud files that contain spiral-shank bolts mixed with other hardware. I need a compact, CUDA-friendly C++ solution that runs directly on an NVIDIA Jetson Orin NX, spots every bolt in each cloud, records the head-center XYZ (and, if practical, its axis direction), and writes the results to a structured XML file. Performance targets • Accuracy: at least 95 % correct identification on the annotated dataset I will supply. • Speed: real-time or near real-time processing on the Orin NX under the standard JetPack image. You are welcome to build on open-source libraries such as PCL, Open3D, Eigen, cuBLAS or TensorRT, provided everything can be compiled through CMake and redistributed without license issues. Deliverables • Well-doc...

...current prompt chains and function calls to pinpoint exactly where Claude Code and Ollama are over-allocating tokens. • Concrete code-level optimisation techniques—prompt refactoring, context window trimming, token-length guards, caching strategies or any other proven methods—that cut the total daily tokens consumed. • Configuration tweaks for both my local machine and the remote H200 server (CUDA, model quantisation, batching, concurrency limits, etc.) so performance improves instead of degrading. • Suggestions for alternative tools, models, or routing logic if replacing parts of the stack would save more tokens than patching them. Deliverables I’d like to see: 1. A brief report that highlights every hotspot you find, shows before-/afte...

I am rapidly up-skilling in CUDA C++ and want an experienced mentor who can walk me through the real-world use of the core foundational libraries—Thrust, CUB, and libcudacxx. My main need is to see clean, well-explained example implementations and concrete use cases rather than abstract theory. Here is what I have in mind: • Short, focused code samples that highlight best-practice patterns in each library (device vectors, reductions, custom kernels, cooperative groups, etc.). • Step-by-step explanations of how these examples map to GPU execution, memory hierarchies, and performance considerations. • Guidance on how to slot each snippet into an existing CMake-based project so I can experiment immediately. I already have a CUDA 12.x toolchain set u...

...statistics and performance monitoring Responsibilities - Designed the architecture for the decentralized AI compute network. - Developed backend services for GPU worker registration, workload management, and job orchestration. - Implemented the communication layer between the central coordination services and distributed GPU workers. - Built AI execution services capable of running workloads on CUDA-enabled GPUs. - Designed the compute accounting and miner reward workflow. - Integrated blockchain functionality for transparent contribution and reward tracking. - Developed APIs for managing users, workers, jobs, compute resources, and rewards. - Implemented real-time monitoring of GPU workers and workload execution. - Containerized AI workers and supporting services using Docker....

...already flashed, updated, and running smoothly—now I need Autoware up and working end-to-end on it. I’m targeting the Perception, Localization, and Path Planning stacks, so the job is to install, configure, and validate those three modules on the board. Here’s what I expect: • Clean installation of the current Autoware version compatible with JetPack on the AGX • All required dependencies (ROS 2, CUDA, TensorRT, etc.) resolved without breaking existing packages • Perception, Localization, and Path Planning nodes launched through reproducible, scripted workflows (Docker or native—your call, as long as it’s reliable) • Demonstration bag file or simulation run proving each module is publishing the right topics at expected rate...

I need an experienced PyTorch/CUDA engineer to squeeze every practical second out of our ROLLCALL Wan 2.2 I2V A14B inference pipeline running on an A100 80 GB while keeping the pictures looking exactly the same. Reducing processing time is the prime objective; any change that simply trades speed for a worse image will be rejected. The first job is a deep profile. Please time each phase separately—model loading, T5/text encoding, VAE, diffusion, decoding, FFmpeg, and all inter-segment overhead—so I can see exactly where the pipeline stalls. From my own sampling it looks as if models may be re-opened for every 5-second chunk, so post-processing and segment overhead are the first areas I’d like you to attack. Once the slow spots are confirmed, create a persistent wa...

...Data visualization - Charting libraries, dashboards, S-curve and trend charts - Preferred Report generation - PDF, Excel, Word, PowerPoint export - Preferred ML / AI Infrastructure PyTorch & Hugging Face - Model fine-tuning, PEFT/LoRA, Transformers library - Preferred Voice / speech ML - STT, TTS, end-to-end voice pipeline development - Preferred GPU infrastructure - Self-hosted GPU servers, CUDA, resource monitoring - Preferred Inference serving - High-throughput model serving, quantization, optimization - Preferred Evaluation - Standard and custom domain-specific evaluation harnesses - Preferred Infrastructure & DevOps Docker - Containerization, multi-stage builds, compose - Preferred CI/CD pipelines - Automated build, test, security scan, deployment workflo...

...that Wan2.2 A14B can generate usable video on an NVIDIA A100 80GB. The previous prototype used RunPod, Python, PyTorch/CUDA, FastAPI/Uvicorn and FFmpeg, but I do not want to continue patching a fragile experimental environment. The engineer may recommend RunPod, Lambda Cloud, CoreWeave, AWS/GCP/Azure or another appropriate GPU provider, but must justify the choice specifically for a large Wan2.2 A14B workload. Required architecture: ROLLCALL Website → API → Persistent Job Queue/Database → Disposable GPU Worker → Wan2.2 I2V A14B → FFmpeg/QC → Object/Persistent Storage → ROLLCALL Requirements: NVIDIA A100/H100-class production inference Wan2.2 I2V A14B Python/PyTorch/CUDA reproducible Docker-based deployment pinned/compatible dependenc...

Senior AI Video / GPU Engineer Needed – Wan2.2 + RunPod A100 + PyTorch/CUDA I need an experienced AI/GPU engineer to finish and productionize an existing AI video-generation backend for a platform called ROLLCALL. This is NOT a website design job. The website is already built. I need someone who specializes in GPU inference, Python, PyTorch/CUDA environments, AI video models, and production API deployment. CURRENT SYSTEM We already have: RunPod NVIDIA A100-SXM4 80GB GPU Wan2.2 I2V A14B Approximately 118GB of Wan2.2 model files already downloaded Persistent /workspace storage Python PyTorch/CUDA FastAPI/Uvicorn worker FFmpeg Existing website integration Existing REST API running on port 3010 Current API routes include: GET /v1/health POST /v1/generate GET /...

...premium platforms). Server Hardware & Environment OS: Ubuntu 26.04 LTS CPU: Intel Core Ultra 9 285K (24 Cores / 24 Threads) GPU: NVIDIA GeForce RTX 5090 (32 GB VRAM, Driver 595.84, 575 W power limit) RAM: 60 GB System Memory Storage: 2 TB Samsung 9100 PRO NVMe SSD Key Responsibilities & Scope of Work Server Setup & Stack Configuration: Complete Ubuntu environment setup optimized for PyTorch, CUDA, and high-vram local inference. Deployment and tuning of ComfyUI, Open WebUI, and n8n. AI Character & Video Production Pipeline: Build advanced ComfyUI workflows leveraging SOTA open-source video/image models. Implement character consistency tools using LoRAs and IP-Adapter/ControlNet pipelines. Integrate lip-sync and dialogue tools (e.g., LivePortrait, MuseTalk, Eleve...

Our NVIDIA CUDA-based servers are online but still missing a production-ready software stack. I need the environment installed, tuned, and kept rock-steady so my research team can start running heavy models without delays. Scope • System setup & configuration – install the latest CUDA toolkit, drivers, cuDNN, NCCL and required OS dependencies, then provision Docker/Container runtime so future upgrades are painless. • Performance optimization – profile current throughput, adjust BIOS, kernel, power and GPU settings, implement mixed-precision or other CUDA tweaks, and document the gains with repeatable benchmarks. • Ongoing maintenance & troubleshooting – create health-checks, monitoring hooks (Prometheus/Grafana preferred)...

I’m expanding a high-performance computing initiative and need seasoned C/C++ talent to squeeze every last cycle out of our code. The core of the work revolves around designing, profiling, and hardening parallel processing routines that will run on Linux/Unix clusters. You’ll be free to choose the right mix of pthreads, OpenMP, MPI, CUDA, or similar frameworks as long as the end result is fast, deterministic, and fully verifiable. Here’s the environment you’ll step into: a mature codebase written almost entirely in modern C and C++, backed by rigorous data-structure design, multithreaded execution, and an SDLC that values deep debugging over quick patches. I handle planning, testing resources, and CI; you focus on writing clean, well-documented modules and d...

...them. • A fraud-detection layer that flags suspicious listings or bidding patterns before a transaction closes. Everything must sit behind a clean, minimal interface—no clutter—because both food and pharmaceutical professionals will be using it every day under time pressure. I’m open on stack, but Python with TensorFlow/PyTorch for the models and a lightweight React or Vue front end makes sense; CUDA optimisation is a plus for the Inception pitch. Acceptance criteria for the MVP: 1. End-to-end workflow: seller upload → AI price suggestion → live auction → automated & manual bids → secure checkout. 2. Latency for price predictions under two seconds on a single Nvidia GPU. 3. Admin dashboard that surfaces fraud alerts and key mar...

...Demo Dashboard Requirements A simple web dashboard should include: Live Camera View AI Detection Overlay Event Alerts Event History Detection Screenshots Camera Management (Basic) Dashboard Statistics A polished UI is preferred but not mandatory for the demo. Preferred Technology Stack Python FastAPI YOLO OpenCV TensorFlow or PyTorch React.js PostgreSQL Docker Ubuntu Linux NVIDIA GPU Support (CUDA) Equivalent technologies are acceptable if performance and scalability are maintained. Future Scope If the demo is approved, the selected developer/team will continue with the complete platform development, including: Face Analytics Safety Analytics Vehicle Analytics Object Analytics Attendance Analytics Multi-location Dashboard Mobile Application Notifications (Email, WhatsApp,...

...on my side. • Deliver the trained model file, an inference script or API endpoint, and a brief report explaining your methodology, final accuracy, and any recommendations for future improvements. I will supply the images and their class labels as soon as we start, and I’m happy to discuss target accuracy or class-imbalance strategies up front. The code should run on a standard GPU instance (CUDA 11.x). Once the model meets the agreed accuracy on my held-out validation set, the project is finished and paid in full....

We are installing EDGE AI devices — mainly NVIDIA JETSON series and higher power units in jails and prisons to detect inmate activity. We need someone familiar with setting up these devices. NVIDIA Jetson™ modules deliver accelerated AI performance at the edge. With the NVIDIA JetPack™ SDK, you can develop and deploy innovative products across industries. The Jetson family uses unified NVIDIA® CUDA-X™ software and supports cloud-native technologies for streamlined AI development and deployment. These prisons and jails often have different routers, securuty and networking. We need experience here to get the message stream from the NVIDIA devices to the Cloud management system.

Looking for an experienced CUDA/C++ developer for short GPU optimization and debugging tasks. Work will require: Direct work on my development environment via AnyDesk/TeamViewer. Tasks include debugging CUDA kernels, fixing tensor indexing issues, improving performance, and optimizing GPU usage.

CUDA Developer Needed – Long-Term Remote Work (AnyDesk/Remote Access) I am looking for an experienced CUDA/C++ developer to help me complete GPU optimization tasks. This is a long-term collaboration opportunity for someone with strong experience in CUDA kernels, PyTorch, and performance optimization. Work setup: Remote work through AnyDesk or another remote desktop tool. You will work directly on my development environment. Tasks involve debugging, improving, and optimizing CUDA implementations. Each task typically requires around 3 hours of focused work. Payment: Rate: $8–$10 per hour depending on experience. Payment will be released after 2 weeks of completed work. This is intended to become a long-term working relationship with regular ta...

I need a Senior Computer Vision and Deep Learning Engineer to build a complete production-ready AI background removal system for AKPRINTHUB, with qualit...third-party paid API should be used. The developer will be responsible for the complete project, including AI model selection/fine-tuning, Python FastAPI backend, GPU optimization, frontend integration with my existing PHP/JavaScript website, testing, bug fixing, deployment and production launch. Required technologies include Python, PyTorch, OpenCV, image segmentation, alpha matting, ONNX/TensorRT, CUDA, Docker and GPU deployment. Complete source code, trained model weights, training scripts, deployment files and documentation must be handed over. Payment will be milestone-based after quality and speed testing against a privat...

CUDA Developer Needed – Long-Term Remote Work (AnyDesk/Remote Access) I am looking for an experienced CUDA/C++ developer to help me complete GPU optimization tasks. This is a long-term collaboration opportunity for someone with strong experience in CUDA kernels, PyTorch, and performance optimization. Work setup: Remote work through AnyDesk or another remote desktop tool. You will work directly on my development environment. Tasks involve debugging, improving, and optimizing CUDA implementations. Each task typically requires around 3 hours of focused work. Payment: Rate: $8–$10 per hour depending on experience. Payment will be released after 2 weeks of completed work. This is intended to become a long-term working relationship with regular ta...

Job Description: I'm running a self-hosted, real-time conversational voice AI pipeline (STT → LLM → streaming TTS over WebSocket) on a RunPod GPU pod (CSM-1B TTS model, using the davidbrowne17/csm-streaming fork with Sesame's Mimi/moshi audio codec, PyTorch ). I need an experienced PyTorch/CUDA engineer to find and fix a reproducible latency bug that's blocking production readiness. The problem: At sentence/turn boundaries in a multi-turn conversation, generation stalls for ~7-8 seconds of dead air. I've already localized this precisely via wall-clock instrumentation: the stall occurs specifically inside Mimi's () call, when encoding a previous turn's generated audio into context — and specifically only when enc...

...privada con mi entorno de agentes desde cualquier dispositivo móvil/celular sin exponer el servidor a riesgos. #### **Requisitos del Freelancer:** * Experiencia sólida con **Ollama**, frameworks de agentes (**OpenClaw**, LangChain/LangGraph o similares) y automatización en **n8n**. * Dominio avanzado de entornos Linux (Debian/Ubuntu/Pop!_OS), gestión de puertos y asignación de recursos GPU (NVIDIA CUDA / ROCm). * Experiencia implementando túneles de red seguros (Reverse Proxies / Tunnels). * Capacidad para entregar código limpio en Python/Bash y documentar brevemente los puertos asignados a cada servicio. Si tienes experiencia montando este tipo de ecosistemas locales eficientes y rápidos, por favor postúlate detalla...

...on a specific AI provider. --- # Required Skills Strong experience with: * Python * FastAPI * PostgreSQL * Redis * Celery * Docker * Linux * REST API Design * Background Workers * Authentication & Authorization * Enterprise SaaS Development Experience with AI model integration is highly preferred. --- Experience with: * AudioCraft * Amphion * OpenVoice * RVC * ComfyUI * GPU Infrastructure * CUDA * RunPod or other GPU cloud providers * AI Media Pipelines --- # Project Status The project architecture has already been designed. Documentation already exists for: * Engineering Constitution * Software Architecture * Database Architecture * Master Roadmap * Development Standards * Security Architecture * AI Platform Architecture we have a "Serverless" NVIDIA acco...

We are building our project AI-native Music SaaS Platform. The platform allows users to generate high-quality songs using AI, including lyrics generation, music composition, ...Mention your experience with Python FastAPI and AI model integration. Please provide links to your GitHub or a project where you built a complex AI pipeline. Skills required Developer Skills • Python (5–10+ years preferred) • FastAPI • PostgreSQL • Redis • Docker • REST APIs • JWT/RBAC • Async programming • System Design • AI integrations • PyTorch • Hugging Face • ACE-Step • JWhisper • FFmpeg • CUDA • GPU Optimization • ComfyUI We would like to humbly request that the same person who has previously worked on ...

...and displays the corresponding information. The initial database will contain approximately **150 registered people**, with the possibility of expanding it in the future. --- # Scope of Work The freelancer is expected to complete the following tasks: ## 1. Prepare the Jetson Platform * Install the appropriate operating system for Jetson. * Configure all required drivers. * Properly install CUDA, cuDNN, TensorRT, and other necessary NVIDIA components. * Configure the Python environment. --- ## 2. Install All Required Dependencies Install and configure all libraries required by the project, including but not limited to: * OpenCV * InsightFace * ONNX Runtime (if required) * FAISS * TensorRT * NumPy * SciPy * PySerial * Any other dependencies required by the project. --- ...

We turn padel courts into smart courts. Two cameras watch each match; players open a shared page afterward to see their highlight c...processing real-time video, so comfort optimizing models for constrained hardware and working with RTSP/GStreamer is valued — the ML is the heart of the role, but you'll build it on the device. You should be at home working on a headless Linux box remotely over SSH (no GUI), moving code and files via scp, and running/debugging everything from the command line. Familiarity with the Jetson/JetPack stack (CUDA, TensorRT) is a plus. The rig is already built and remotely accessible; a full technical handoff goes to shortlisted candidates. To apply: share relevant CV/ML work and a short note on how you'd approach the detection → track...

...It's currently failing on newer NVIDIA GPU hardware due to CUDA/driver/dependency compatibility issues. I need someone to: Get the pipeline running cleanly on the target GPU environment (debugging CUDA/driver/dependency conflicts) Run evaluation and confirm standard output metrics are generated correctly Deliver clean, runnable source code with basic documentation and a demo of it working Requirements: Strong PyTorch experience Experience with CARLA simulator or similar simulation environments Comfortable debugging CUDA/driver/dependency issues, ideally with newer GPU hardware Access to a compatible GPU environment (local or cloud) Please tell me: Your experience with CARLA and/or autonomous driving models Your experience with CUDA/GPU compatibility...

I need a production-ready GPU instance on E2E Cloud that runs the Qwen vLLM model for object detection. The service must accept image files through a simple REST endpoint, run inference, and return bounding boxes and labels—stable enough to handle 200 k – 500 k calls every day. Here’s the workflow I have in mind. You’ll provision a suitably powerful E2E Cloud VM, install CUDA, cuDNN, vLLM, pull the latest Qwen checkpoints, and wire everything together with a lightweight Python server—FastAPI is my usual choice, but feel free to suggest an alternative as long as it stays lean and well-documented in OpenAPI/Swagger. Throughput matters: I’m aiming for at least 2–5 sustained requests per second, so you can rely on batching, concurrent workers, ...

... • A fraud-detection layer that flags suspicious listings or bidding patterns before a transaction closes. Everything must sit behind a clean, minimal interface—no clutter—because both food and pharmaceutical professionals will be using it every day under time pressure. I’m open on stack, but Python with TensorFlow/PyTorch for the models and a lightweight React or Vue front end makes sense; CUDA optimisation is a plus for the Inception pitch. Acceptance criteria for the MVP: 1. End-to-end workflow: seller upload → AI price suggestion → live auction → automated & manual bids → secure checkout. 2. Latency for price predictions under two seconds on a single Nvidia GPU. 3. Admin dashboard that surfaces fraud alerts and key marke...

Tired of boring, soul-crushing jobs? So are we. We're not just another AI shop – we're a people-first business on a mission to solve real-world problems, from removing CO₂ from the atmo...to Include: · CV/Resume highlighting Python AI production experience · GitHub/Portfolio with AI deployment projects · Brief proposal on how you'd approach this · References from previous clients/employers Subject Line: "Python AI Engineer - [Your Name]" --- IMPORTANT: · Do NOT apply if you only have Jupyter/R&D experience – this is production. · Do NOT apply if you cannot work with GPUs and CUDA. · Do NOT apply if you cannot commit to the 7–10 day timeline. · Be ready to start immediately. ...

...NVIDIA GPU Operator, NVIDIA Network Operator, CNI, CSI, and similar Kubernetes ecosystem tools. * Experience with job scheduling systems such as Slurm. * Strong Linux system administration skills. * Proficiency in scripting and automation using Python and Bash. * Experience with observability and monitoring platforms such as Prometheus, Grafana, and Loki. * Knowledge of GPU architectures, NVIDIA CUDA, NCCL, and AI/ML infrastructure is a strong advantage. * Strong troubleshooting and root-cause analysis skills with the ability to analyze logs, metrics, and system performance data. * Excellent communication, collaboration, and problem-solving abilities. Preferred Skills * Large-scale Kubernetes cluster operations. * AI/ML infrastructure and GPU cluster management. * Infrastructur...

I need the open-source InsightFace face-swap model installed and fu...on my own Ubuntu server or an inexpensive rented GPU instance. • A test run that proves the pipeline can process sample footage and photos without quality loss, including a short 1080p clip and a few stills. • Guidance on how to queue jobs, control face detection, and tweak blending parameters so each swap costs only a fraction of a cent in compute time. Please make sure any dependencies—CUDA, PyTorch, FFmpeg, and InsightFace weights—are pulled automatically or documented precisely so I can rebuild the environment later. If you prefer another lightweight orchestration tool, I’m open to it as long as the result remains simple to maintain. Once everything runs end-to-end on my m...

...images for Tenstorrent AI ASICs to expand our hardware ecosystem beyond current GPU deployment - Migrate Python code and VLLM implementations to new VLLM images and adapt them for specific GPU cards Required Qualifications: - Proven experience with large language model optimization techniques - Strong understanding of transformer architectures and attention mechanisms - Proficiency with PyTorch, CUDA, and GPU optimization techniques - Experience with vLLM, FlashInfer, or similar inference optimization frameworks - Familiarity with Docker containerization and GPU workload management Preferred Qualifications: - Experience with Claude Code Max (will be provided if needed) - Previous experience with Gonka or similar decentralized AI networks - Background in competitive ML or distri...

... • Helm or ArgoCD handles drift management so that the desired state remains in sync. • At runtime, the service must auto-scale both CPU and GPU requests, proving horizontal and vertical elasticity. • Token or currency budgeting is tracked per request so I can see live cost data and later hook it into FinOps dashboards. Inside the cluster • Nvidia AI Enterprise stack (Triton, TensorRT, CUDA) installed via the operator. • LLM endpoints wrapped with LangChain agents that can call external tools. • A vector database (FAISS, Milvus, or anything OSS) stores embeddings for RAG. • One sample agent demonstrates question-answering over the vector store; another shows a multi-tool plan/act loop. What I need from you 1. Terraform or simil...

I already have a RunPod GPU instance waiting and need a clean, production-ready setup of an open-source large language model that will power a technical-support chatbot. Here’s what has to happen: • Provision the RunPod container (CUDA drivers, Python, Hugging Face Transformers, LangChain, bitsandbytes, etc.) and verify the GPU is fully utilized. • Load and configure an open LLM (I’m leaning toward Llama-2 or similar; suggestions welcome) with the right quantization to fit the available VRAM while maintaining answer quality. • Expose a secure REST or WebSocket endpoint—FastAPI is fine—so my front-end can pass user queries and receive streamed responses. • Add basic retrieval-augmented generation hooks so the bot can reference m...

...current setup, identify the bottlenecks, and implement practical optimizations so the model replies faster and runs leaner without harming output quality. We’ll decide together which metric (latency, memory footprint, throughput) will matter most once you’ve profiled the system, but the immediate goal is measurable, benchmarked improvement. Expect to work with PyTorch, Hugging Face Transformers, and CUDA; if you prefer alternatives such as TensorRT, ONNX, or quantization/pruning frameworks, feel free to propose them. Deliverables • A concise optimisation plan after your initial audit • Updated code, scripts, or checkpoints reflecting the changes • A before-and-after benchmark report showing the gains I’m paying hourly and will fund mileston...

...help you hit the ground running. What I expect from you 1. Build the Python rule services, wired into the message queue layer. 2. Create REST/JSON APIs so external apps can trigger, modify, and monitor rules. 3. Provide Terraform/Ansible (or comparable) scripts that spin up VMs, schedule containers, and deploy updates with zero downtime. 4. Optimise for GPU tasks when a node advertises CUDA. 5. Document the entire flow clearly: architecture diagrams, setup steps, and example calls. 6. Deliver a full test suite that covers rule correctness, scaling behaviour, and fail-over scenarios. Acceptance is straightforward: I’ll spin up the provided scripts on a fresh cloud account, run the tests, and verify that a sample rule set executes end-to-end across at least two ...

...augmentation and dataset pipelines comparable to tf.data. • Model Evaluation – training loops, metric tracking, checkpointing and model save/load logic. Runtime expectations The library will be used in standalone fashion; it only needs to expose a public API that any C# solution can reference. Low-level compute may target CPU initially, but the architecture should stay extensible enough to plug in CUDA, DirectML or other accelerators later. Deliverables 1. Source code with clear namespace organisation and XML documentation. 2. A sample project that trains and evaluates at least one CNN on MNIST using only the new library. 3. Step-by-step build instructions and a short design document outlining graph execution, back-propagation implementation and extension p...

...Python 3.11, FastAPI, SQLAlchemy + Alembic, PostgreSQL, Redis, WebSocket • AI layer – YOLOv8 (people detection / counting), DeepSORT (tracking), RetinaFace + InsightFace-ArcFace (face detection / recognition), FAISS for vector search, OpenCV to pull RTSP streams • Frontend – 15 with TypeScript, Tailwind CSS, Recharts for dashboards • Infrastructure – Docker, docker-compose, NVIDIA CUDA runtime for GPU inference, Nginx reverse proxy Backend reliability and performance sit at the top of the priority list, so I expect an asynchronous FastAPI setup, connection pooling, health checks, graceful shutdown, and clear separation between I/O-bound and GPU-bound tasks. Deliverables should include: 1. Well-documented repo (or mono-repo) layout f...

...powered by an RTX 5090 and want to turn it into a reliable, local playground for the latest stable releases of Stability Matrix, Stable Diffusion, MidJourney, ComfyUI, and Kohya. I need more than a basic installer—I’d like the full environment properly configured, GPU-accelerated, and ready to create images the moment we finish. You’ll connect through a secure remote session, handle every dependency (CUDA, Python, libraries, checkpoints, etc.), and tune paths so each tool cooperates without conflicts. Where optional models or extensions improve usability, feel free to suggest and add them as long as they remain stable. Deliverables • All five applications installed in their most recent stable versions • Verified ability to generate a sample image...

I am looking for an expert to develop a high-performance, 8-channel automated farming system for the 3D tactical shooter Delta Force. My requirements: Technical Skills & Experience: - Proven experience in real-time computer vision (YOLO, TensorRT) with batch inference and low latency on RTX 4070Ti/4060Ti GPUs. - Strong proficiency in C++ (or high-performance Python with CUDA/C++ backend) and PCIe capture card integration. - Familiarity with KMBOX Net HID-level mouse/touch emulation and anti-cheat evasion (Tencent ACE or similar). - Experience designing centralized dashboards and robust fail-safe management for 24/7 operations. Core Deliverables: - Real-time detection (loot, enemies, extraction points) and UI state recognition via 8x 1080p 60Hz streams. - Visual navigation usi...

I need an Ubunt...so I can run machine-learning workloads. The job is purely the installation and configuration of everything required for the guest OS to recognise and fully utilise the GPU (nvidia-smi must work). Key points • VMware host is already installed and stable; you decide whether PCI passthrough, vGPU or another VMware feature makes most sense. • Inside the VM I will need the correct NVIDIA driver, CUDA toolkit and cuDNN ready for TensorFlow or PyTorch later on. • Once complete I should be able to launch a quick test script that confirms the GPU is visible from within Ubuntu. Deliverable A step-by-step session (screenshare or detailed command log) that leaves the VM running and the GPU operational, plus any commands I can rerun if I ever have to r...

...Video Processing * FFmpeg integration required --- ## File & Project Structure Each project should store: * Audio file * Lyrics * Scene data (JSON) * Generated images * Generated clips * Final output --- ## Advanced Features (Preferred) * Beat-synced cuts and transitions * AI-assisted prompt generation * Style consistency across scenes * Character continuity (optional) * GPU acceleration (CUDA support) --- ## Deliverables * Fully working desktop application * Clean, maintainable Python code * Installation/setup instructions * Ability to run locally without cloud dependency (preferred) --- ## Notes for Developer * This is not a simple generator — it is a **production tool** * User control over scenes is critical * Performance and stability are important * M...

...Ubuntu and I need the open-source WAN Video 2.7 stack fully installed and running on it. You will connect over AnyDesk, handle the complete setup, pull all required libraries and system dependencies, and verify the application launches cleanly with GPU acceleration enabled. During the session I will stay online to provide root access and restart the box if needed. Please be comfortable working in a CUDA-centric environment, compiling from source when binaries are not available, and troubleshooting driver or codec conflicts that sometimes appear on DGX hardware. Acceptance criteria • WAN Video Open Source 2.7 starts without errors and streams a test feed. • All supporting packages and services are documented in a short README so I can replicate the build later. &...

...Long-term work possible if successful Scope of Work: - Diagnose deep learning pipeline issues - Fix model execution errors - Debug training / inference workflow - Resolve dependency or environment conflicts - Optimize pipeline stability - Ensure end-to-end execution works correctly - Provide brief documentation of fixes Technical Stack: - Python - PyTorch / TensorFlow - HuggingFace / Transformers - CUDA / GPU acceleration - Docker / Linux environment - API integration & Data preprocessing pipeline Requirements: - Strong experience in Deep Learning production workflows - Experience debugging complex AI pipelines - Comfortable working under urgent timelines and ability to start immediately Timeline: Start: Immediately. Expected turnaround: 24–48 hours. Proposal Requi...

I’m building a camera-based ... log every reading with a timestamp, and trigger a visual or audible alert whenever negative emotions are detected repeatedly within a short window. A lightweight dashboard served with either Streamlit or Flask will let me: • watch the annotated video feed • view rolling emotion statistics and charts • review and download the timestamped log of events and alerts Optimisation for Jetson (CUDA, cuDNN, TensorRT where appropriate) is essential, and the finished app should launch from a single command, open the dashboard in a browser, sustain real-time performance, and shut down cleanly. Please keep the code modular and well commented so I can retrain or swap models later and, if convenient, provide a Dockerfile or setup script ...

...personalized address generator written in C/C++ and CUDA. The tool should have the following features: • Read any number of prefix-suffix patterns from the `` file (example format: `Taaa*1111` or `Tbbb*222`). • Launch a GPU kernel to continuously generate wallet addresses and compare each address with all patterns. If a match is found, write the matching address and its private key to disk. • Fully utilize GPU performance, achieving the same speed as my current test version (approximately 8 billion addresses per second). Please display a "addresses per second" counter in real-time during program execution. • Generate a plain text log file recording key events: startup time, device information, running hash rate snapshots, and each match found. ...

... "Zero-Shot" Virtual Try-On pipeline into an existing Flutter/Python e-commerce stack. Technical Stack Requirements AI/ML: Experience with IDM-VTON, Cat-VTON, or OOTDiffusion. Mastery of Stable Diffusion (ControlNet/IP-Adapter) is mandatory. Computer Vision: Expertise in MediaPipe or OpenPose (pose estimation) and DensePose (surface mapping). Backend: Python (FastAPI/PyTorch), gRPC/REST, and CUDA optimization. Frontend Integration: Flutter (Dart) for image handling and state management. Key Deliverables The "Zero-Retrain" Pipeline: A model that accepts a flat garment image and a user photo to produce a drape-accurate result without per-SKU training. Latency Optimization: Implementation of TensorRT or AITemplate to bring inference time under 3 seconds on ...