Hluboké učení Jobs

I need a fully-async Python 3.11+ pipeline that turns a live phone call into a smooth, sub-2.5 s p50 conversational loop. Here’s the flow I have in mind: a Twilio SIP trunk delivers the RTP stream to a WebSocket bridge; LiveKit Agents SDK manages the media session; Deepgram Nova-3 handles streaming STT; the running transcript feeds Claude (with tool-use enabled); Claude’s text comes back out through ElevenLabs Flash v2.5 for TTS and streams to the caller in real time. What I need from you is a working reference implementation, instrumented and tuned so I can see latency at each hop and fine-tune Voice Activity Detection thresholds. The code must retry transient errors, log everything that matters, notify the caller gracefully on trouble, and, if the Sonnet tier fails mid-call...

I need a robust computer-vision pipeline that can automatically spot and label consumer-goods products in high-resolution images captured on our manufacturing line. The goal is to distinguish each finished item from background equipment, operators, and any other visual noise so that we can feed the detections into our downstream QA and inventory systems. You will start with raw JPEGs taken under factory lighting, then design, train, and validate an object-detection model—PyTorch or TensorFlow is fine—capable of achieving consistent, real-time performance on an NVIDIA GPU. If you prefer a different framework, let me know why; I’m flexible as long as the final solution is easy for my engineering team to maintain. Deliverables • Annotated sample dataset (in COCO or ...

K zobrazení podrobností se zaregistrujte nebo přihlaste.

I want to build a robust machine-learning pipeline that can reliably predict when a user is likely to log on again. The core need is a production-ready model—deep-learning is welcome where it adds value—that captures behavioural signals and translates them into accurate logon-probability scores. You will work with the raw system logs I can provide (time-stamped events, account metadata and any other fields you advise extracting). Starting from exploratory data analysis, we will move through feature engineering, model selection, hyper-parameter tuning and final evaluation. I am especially interested in interpretable insights alongside raw accuracy, so attention to explainability techniques such as SHAP or LIME is appreciated. Deliverables • Clean, well-commented Python ...

I am preparing a new round of model training and need a senior-level AI developer who can take full ownership of the data-preparation stage—especially data annotation. The core of the job is to design and implement a robust annotation workflow for a sizeable corpus of text data, then feed that clean, well-labeled material back into the training loop. You should already have hands-on experience setting up annotation guidelines, managing annotators or automation tools, and integrating the resulting labels into a machine-learning or deep-learning pipeline. Familiarity with popular NLP libraries (spaCy, Hugging Face, TensorFlow, PyTorch, etc.) will be essential, as the final objective is to boost downstream model performance by improving label quality and consistency. Deliverables &b...

Saya ingin mulai menghasilkan uang melalui blog berisi ulasan produk kecantikan. Untuk langkah awal, saya membutuhkan bantuan menyiapkan konten review yang menarik, informatif, dan ramah SEO sehingga mudah ditemukan pembaca sekaligus berpotensi mendatangkan affiliate‐sales. Apa yang saya butuhkan: • Riset singkat tentang tren dan kata kunci relevan di niche kecantikan. • Draft ulasan produk yang orisinal—menjelaskan manfaat, cara pakai, kelebihan, kekurangan, dan rekomendasi jujur. • Gaya bahasa ringan tetapi tetap profesional, mudah dipahami pembaca Indonesia. • Struktur artikel lengkap: judul catchy, meta description, heading teratur (H1–H3), call-to-action, serta saran link internal/eksternal. • Penyematan data teknis (harga, varian, bahan u...

I have collected hundreds of standard photographs of hands in many poses and from many people. Nothing is uniform: skin tones, hand sizes, lighting, backgrounds, and even finger counts may vary. The goal is to turn this dataset into a practical analysis tool that can: • calculate the flexion/extension angle at every finger joint in each image • flag when a finger or phalanx is completely absent so that it can be recorded automatically I expect you will combine classical computer-vision preprocessing with a deep-learning model (PyTorch or TensorFlow are fine) and possibly leverage pose-estimation libraries such as MediaPipe or OpenCV for landmark detection. Accuracy matters more than perfection in lighting or background, so robust data-augmentation and domain-adaptation tech...

We want to use the existing data and use other concepts for data augmentation compared to what we have now. Then we want to retrain the machine learning model.

I have a growing archive of high-resolution intraoral photos and I need an AI mechanism that can automatically locate every tooth in each image, assign the correct FDI or Universal numbering, and clearly classify each tooth type (incisor, canine, premolar, molar, wisdom). You are free to decide whether a pre-trained or custom-built model is best; accuracy and speed matter more to me than the underlying brand of framework. Python with PyTorch or TensorFlow, OpenCV for preprocessing, and a clean REST or CLI inference interface would suit my workflow, but if you have a more elegant stack, I’m open. Key deliverables • A trained model capable of processing new intraoral 3d scans and returning JSON with tooth positions, numbers and classes. • Lightweight script or API e...

K zobrazení podrobností se zaregistrujte nebo přihlaste.

I need a complete, ready-to-run video analytics tool that flags car-collision accidents in city-surveillance footage. The workflow must stay simple for the end user: they upload a clip, hit “Analyze,” and, within seconds, the screen returns something like: Accident Detected: YES Severity: HIGH (low | medium | high) Confidence: 92 % Time-stamp: 00:12 Key points to build in • Scope of detection: only car collisions; no pedestrian or bicycle tracking at this stage. • Source footage: city CCTV style feeds (fixed street-level cameras). • No real-time push notifications are required—the result can appear once processing is finished. I will rely on you to select or curate a robust, publicly available dataset (or a combination of datasets) that...

I am expanding my software-engineering agency into the US, Canada, and European markets and I need a confident voice on interview calls who can represent our senior developers. So US, Canada, Australia Citizens are prefered. When a potential client schedules a technical interview—whether the topic is JavaScript micro-services, Python data pipelines, Machine Learning with PyTorch, or the finer points of Software Architecture—you will join the call, introduce yourself on behalf of the engineer, and guide the conversation toward closing the engagement. Because every prospect is different, I am looking for someone already comfortable in the technology sector. Familiarity with Software Architecture is essential; being able to reference image processing, deep-learning or NLP concept...

I need a rock-solid, real-time player tracking module for football matches that guarantees the ID assigned to each athlete at kick-off never changes until the final whistle. Right now, our OpenCV–TensorFlow–YOLO pipeline sometimes swaps or loses IDs when athletes overlap, leave the frame briefly, or the camera angle shifts, and that ruins every speed, distance, position, and heat-map metric we generate. Key requirements • Sport: football. • Camera setup: five or more synchronized feeds. • Existing stack: OpenCV, TensorFlow, YOLO – your solution must plug into this environment. What I expect 1. A multi-object tracker with integrated re-identification that preserves the same unique ID through occlusion, crossings, short disappearances, or camera changes...

I’m producing a live-action music video built around a love-and-relationships storyline and want to amplify it with cutting-edge AI work. Principal footage will already be shot (approx. 3 min), and the task now is to layer in artificial-intelligence magic that elevates the emotional arc without losing the organic feel of the performances. The two core enhancements I need are: • Special effects and cinematic filters that accentuate mood shifts (think colour-graded dream sequences, particle light flares, etc.). • Facial recognition and tasteful face substitution for brief flashback moments where the same actors appear younger/older or in imagined scenarios. Smooth integration is essential—skin tones, lighting and lip-sync have to stay believable. Adobe After Effec...

I'm seeking an AI/ML programmer to develop an image classification model specifically for fingerprint recognition. Key requirements include: - Expertise in image recognition and classification - Experience with deep learning frameworks (e.g., TensorFlow, PyTorch) - Strong background in handling and processing image datasets - Knowledge of security and privacy considerations in biometric data Ideal candidates will have a proven track record in similar projects and the ability to deliver accurate and efficient models.

I have just 700 indoor photos that must be labelled for an my project upcoming object-detection model. Every visible person and every seat that is currently occupied needs its own bounding box, and because the network will train on oriented-bounding-box (OBB) data, the rectangles have to follow the exact rotation of the body or seat—especially when someone is leaning. Data is of infrared camera of a cinema check samples , please note there can be many people in one image i want each to be marked properly You will work inside the simple browser-based annotation tool I built for this purpose or any other annotation tool you like. drag the box corners, spin the angle handle where needed, hit save, and move to the next frame; the app tracks progress automatically so nothing is missed....

I want to put an AI tutor in every algebra student’s pocket. To make that happen, I need you to design and code a Retrieval-Augmented Generation pipeline that can
understand a question, pull the most relevant snippets from my curated algebra textbooks, formula sheets, and problem sets, then explain the answer step-by-step with rock-solid reasoning. Here is what success looks like from my side: • A working RAG stack (embeddings → vector database → LLM) wired together with modern tooling such as LangChain, LlamaIndex, or an approach you can justify. FAISS, Pinecone, Weaviate, or similar are fine for the vector layer; OpenAI, Claude, Llama 2/3, or comparable models may drive generation. • Retrieval tuned for high recall and low latency—students should see a...

K zobrazení podrobností se zaregistrujte nebo přihlaste.

Hybrid Artificial Intelligence System for Autonomous Drone Detection Using Computer Vision and RF Signal Analysis The idea is to build AI detection system based on hybrid method (Computer vision and RF signal) and present results :Accuracy, Precision, Recall and False Alarm Rate Algorithms: Computer vision: YOLOv8 (Object Detection) RF analysis: CNN for Spectrogram analysis Merge in : Rule-Based Fusion • or Neural Network Fusion Dataset : Vision Datasets: • Anti-UAV Dataset • VisDrone RF Datasets: • DeepSig RadioML Dataset • Custom SDR Capture (HackRF / RTL-SDR) Budget : 150$

We are looking for a talented frontend developer to bring our Figma designs to life by building a visually stunning, high-performance website using HTML, CSS, and JavaScript. This project involves translating Japanese content, so proficiency in Japanese is required. The ideal candidate will have a strong eye for design and be able to accurately replicate layouts while enhancing them with smooth animations and professional styling. Responsibilities: Convert Figma designs into clean, responsive HTML/CSS/JavaScript code Ensure pixel-perfect implementation aligned with the original design Translate Japanese content accurately into the website Implement modern UI animations and interactions Optimize for performance and cross-browser compatibility Requirements: Proven experience in frontend deve...

I have just 700 indoor photos that must be labelled for an my project upcoming object-detection model. Every visible person and every seat that is currently occupied needs its own bounding box, and because the network will train on oriented-bounding-box (OBB) data, the rectangles have to follow the exact rotation of the body or seat—especially when someone is leaning. Data is of infrared camera of a cinema check samples , please note there can be many people in one image i want each to be marked properly You will work inside the simple browser-based annotation tool I built for this purpose or any other annotation tool you like. drag the box corners, spin the angle handle where needed, hit save, and move to the next frame; the app tracks progress automatically so nothing is missed....

The project centers on building a production-ready TensorFlow 2.x model that classifies tabular data delivered to us through an internal API. I have the API specifications and sample payloads ready; you will turn those streams into a clean training pipeline, engineer the right features, and iterate until the classifier meets our performance targets in real-world tests. Scope of work • Data pipeline – pull the API data, handle preprocessing, and produce TensorFlow-friendly datasets for train/val/test splits. • Model development – design, train, and tune a deep learning architecture suitable for tabular inputs (e.g., wide & deep, Transformer, or other proven structures). • Optimization – experiment with hyperparameters, regularization, and callback...