Anomaly detection jobs
disapprove a link in , it also deletes link from master database disapprove links is still having trouble detect body text of main channel page for 3. currently disapprove links is still having trouble to save links correctly into filteration_words_one.csv. Sample saving format is for link is rupannnanasei
...Articles may use AI-assisted summaries, but: No auto-publish No spinning Human final edits required Clear internal workflow supporting this 3. RSS → Draft Workflow (Google-Safe) I will be sourcing content via RSS feeds. You must configure: RSS feeds importing only into Draft / Pending Review No automatic publishing Source allowlist (approved feeds only) Post/day limits Duplicate title detection Category-based limits if needed You must explain which plugin(s) you will use and why. 4. Content Quality Guardrails To avoid spam and policy issues: Minimum content length enforcement (after human edit) Ability to manually reject or rewrite drafts No thin, empty, or placeholder articles published Clear separation between: Draft Pending Review Published 5. Cate...
I'm seeking an experienced AI developer to create a computer vision model focused on detecting people. The model will need to function effectively in both indoor and outdoor environments. Key Requirements: - Primary function: Object detection with a focus on people - Adaptable to both indoor and outdoor settings - High accuracy and reliability Ideal Skills and Experience: - Expertise in AI and machine learning - Strong background in computer vision, particularly in object detection - Experience with datasets and training models for varied environments - Proficiency in programming languages such as Python, and familiarity with libraries like TensorFlow or PyTorch Please provide examples of similar projects you've completed.
...in image recognition. - Strong experience with Python and API development. - Familiarity with automated visual inspection systems. - Prior experience in the aerospace industry is a plus. AI-Based Crack Detection Using Camera and Image Processing This project aims to develop an AI-based visual inspection system that uses camera images to automatically detect cracks and surface defects on aircraft metal panels. A deep learning model is trained on labeled images of surface cracks and is used to analyse new images captured by a camera or uploaded by the user. The system outputs the detection result in the form of: • the location of detected cracks on the image, and • a final inspection decision for each image (crack detected or no crack detected). The purpose o...
...Configure multi-platform auto-upload through Comet, mapping captions, thumbnails, tags, and any platform-specific metadata. • Create scheduling rules so each post goes live at the time I set—daily, weekly, or according to custom calendars. • Build a distribution workflow that queues content, validates it against each API’s limits, and triggers retries on transient errors. • Add an error-detection layer with clear logging and alerts so I can see failures, root causes, and automatic retries in one place. • Deliver deployment scripts and concise documentation so I can run the system locally or inside our existing cloud environment. The finished deliverable should include: 1. Source code and configuration files. 2. A short architecture diagr...
...Priority bugs within our business management platform, CompanySync. Our application is built and hosted on the Base44 platform. You will be working directly within the Base44 editor to debug React components, Node.js serverless functions, and complex AI integrations. Critical Priorities: Fix AI Voice Interruption (BUG-001): Resolve issues with Gemini Live Voice where interruption and Voice Activity Detection (VAD) are failing. Twilio Inbound/Outbound (BUG-013/014): Fix TwiML handlers for inbound AI voice calls and SMS responses for our virtual assistants. Google Calendar Two-Way Sync (BUG-003): Troubleshoot and resolve domain verification and webhook watch channel issues in the Google Cloud Console. Data Deduplication (BUG-004/005): Fix race conditions creating duplicate comp...
I have a commercial building...generate coordinated floor plans, sections, schedules and construction details without extra patch-up work. Pay attention to correct story settings, composite build-ups, library part selection and intersection priorities so clashes surface early. Deliverables • Native Archicad PLN (version 26 or newer) with all attributes, layers and pen sets neatly organised • IFC export suitable for clash detection in Solibri / Navisworks • Two layout sheets showing general plans and a live section to confirm the model reads correctly I’ll be available for quick clarifications and weekly screen-share reviews; happy to provide redlines as you progress. Let me know your estimated turnaround and any similar commercial projects you have m...
...enables upload of DICOM files, automatic anonymization (strip all metadata/identifiers), AI-driven detection & annotation of pathologies (caries, periodontal disease, bone loss, cysts, etc.), visual overlays, confidence scores, and report generation—all with strict patient privacy, no storage of originals, and human oversight required. Key Requirements: • Clean React/ frontend with drag-and-drop upload, DICOM viewer (e.g., ), annotation overlays & heatmaps. • Python backend (FastAPI preferred) + secure auth, encrypted file handling, and cloud storage (AWS S3/GCP). • PyTorch/TensorFlow ML models (fine-tune YOLO/U-Net/MONAI on open dental datasets) for multi-label detection/segmentation. • Mandatory: Full anonymization on upload (pydico...
...is happening so I can flag hotspots and act quickly. Scope of work • Acquire and pre-process the latest Sentinel imagery for my specified AOIs in the tropics (cloud masking, atmospheric correction, seamless mosaicking). • Run an automated change-detection routine that distinguishes new clear-cuts from seasonal or spectral noise. • Output intuitive products—shapefiles, GeoTIFFs, and a simple dashboard or web map—that visualise forest-loss polygons with dates, area statistics, and confidence scores. Acceptance criteria • Detection accuracy is demonstrably high when validated against at least two recent ground-truth examples I will supply. • Processing chain is reproducible in Python or Google Earth Engine, with commented c...
I need an expert to improve the accuracy of a histopathologic cancer detection model. The current model needs enhancement, and I prefer using algorithm enhancement for this task. Key Requirements: - Improve the model's accuracy in detecting cancerous tissues. - Use advanced techniques and methodologies for algorithm enhancement. Ideal Skills and Experience: - Expertise in machine learning and deep learning - Strong background in medical image analysis - Experience with histopathological images - Proficiency in Python and relevant libraries (TensorFlow, Keras, PyTorch) - Familiarity with model evaluation and performance metrics Please provide examples of similar work and a detailed approach to how you would tackle this project.
...slot straight into the current BIM model (Revit 2024) and align with the latest architectural set. • Load calculations, equipment schedules and single-line diagrams sized for cost-effective installation. • A concise value-engineering report highlighting where we can trim material and labour expenses, together with any alternate product suggestions that preserve maintainability. • Clash detection reports after each coordination round with structural and architectural disciplines. • Attendance at two virtual coordination workshops with the wider design team (GMT+3). Acceptance criteria • All documents must meet local code (ASHRAE + applicable municipal standards) and the landlord’s shopping-centre technical guide. • Drawings delivere...
CLOUD-NATIVE SUPTECH PLATFORM Developer Implementation Master Specification (PoC Build Pack) 1. PURPOSE OF THIS DOCUMENT This document provides the complete technical build specification for developers implementing the Cloud-Native Regulatory System (CNRS) — a supervisory technology platform for central bank supervision, risk analytics, compliance monitoring, early-warning systems, fraud detection, and stress testing. The platform is cloud-native (AWS), modular, AI-enabled, and supports regulatory analytics aligned with Basel III, AML/CFT, and Risk-Based Supervision frameworks. 2. SYSTEM OBJECTIVES The system must: Ingest regulatory data from financial institutions securely Validate, store, and process supervisory datasets Compute prudential ratios and compliance rules Detec...
I already have a working object-detection pipeline written in Python, and I now need that same logic moved into a cleaner, better-structured Python codebase that’s easy to maintain and integrate into a larger application. Think of it as a conversion/refactor: exact same model, exact same results, but with modern syntax, clear separation of concerns, and thorough inline comments. You’ll start from my original scripts and checkpoints, preserve every bit of accuracy, and hand back a fully functioning module (including a simple demo script) that can be installed with pip-installable requirements. Feel free to streamline library calls—TensorFlow, PyTorch, OpenCV, or whatever is currently in place—so long as the final inference output matches the reference I provi...
Multilingual Speech + LLM Integrated Chatbot • Built a multilingual chatbot that understands and responds to users in their native language using real-time speech. • Used automatic language detection, speech-to-text, translation, and LLMs to generate accurate responses. • Created a 3D animated avatar that speaks responses with synchronized voice for a natural experience. • Designed modular and scalable pipelines for speech processing, translation, and AI integration.
I need an expert to develop embedded algorithms for our OWLY predictive maintenance platform. The platform uses fused vibration and acoustic sensing to detect early equipment faults in home appliances, specifically washing machines, dryers, and refrigerators. Key Requirements: - Convert sensor data into actionable equipment health diagnostics. - Focus on early fault detection for motors, bearings, and gearboxes. Ideal Skills and Experience: - Strong background in embedded systems and algorithm development. - Expertise in signal processing and sensor data analysis. - Experience with home appliance maintenance or diagnostics. - Familiarity with machine learning techniques is a plus.
...looking for a senior-level NestJS developer to build a robust, production-ready car data import engine for our vehicle marketplace. This isn't a simple "upload and save" task; we require a sophisticated streaming pipeline capable of processing massive datasets (CSV, XML, JSON) with minimal memory footprint and high reliability. Core Task Build a POST /imports/cars endpoint that: Automatic Format Detection: Handles multipart/form-data and identifies the file type (CSV, XML, or JSON) programmatically. Stream-Based Processing: Processes data using Node.js Streams / AsyncIterables. The application should never load the full file into RAM. Data Pipeline: Implements a clean Parse → Validate/Normalize → Map → Persist flow. Smart Idempotency: Ensures no...
...new clients per week (typical contract £500–£1,000) --- Scope of Work (Detailed) Phase 1 – System Setup You will help configure: Email setup Google Workspace or Microsoft 365 (SPF, DKIM, DMARC) CRM HubSpot Free or Sales Hub Starter only Contact, company, deal pipelines Custom fields for compliance justification and opt-out tracking Basic automation Sequences (rate-limited) Reply detection Auto-pause on reply HubSpot Professional / Enterprise is explicitly NOT required. --- Phase 2 – Prospect Identification Support You will assist with: Identifying organisations (not bulk email lists) in: Care sector Managing agents Commercial landlords Geographic focus: Merseyside Sourcing work email addresses only (no personal emails) --- P...
I have a small batch of images—fewer than one-hundred—that need clean, consistent object-detection labelling. For each image you will draw tight, non-overlapping bounding boxes around every instance of the target classes I will supply once we start. Accuracy matters more than speed; missed objects or sloppy boxes will be rejected. Preferred workflow is any modern tool that can export to COCO JSON or Pascal-VOC XML, as these formats plug straight into my training pipeline. If you normally use LabelImg, CVAT, Supervisely, or similar, that’s perfect. Deliverables • Annotated dataset in COCO JSON or Pascal-VOC XML (your choice, just stay consistent). • A quick text report summarising class counts and any edge cases flagged during labelling. I wil...
I’m sitting on a few hundred PDF invoices and need an automated way to spot any that were issued twice. Because our numbering scheme uses custom prefixes and special characters, I’d like the detection to rely solely on customer-specific data: the vehicle card number, the number of litres dispensed, and the exact date-time of tanking. Here’s what I’m after: an AI-powered script or lightweight app that scans every PDF, extracts those three fields with high accuracy, and then flags, groups, and reports any duplicates it finds. A clear CSV or Excel file listing each suspected duplicate (together with a confidence score and page reference) will be enough for me to review and act on. Acceptance criteria • All PDFs processed automatically—no manual ...
I have a set of user-generated clips that need to be turned into a polished, professional-looking advertisement. The entire workflow should be handled through AI: smart scene detection for automated cuts, tasteful AI-generated filters and effects that elevate—but never overpower—the footage, and perfectly synced automated subtitles and captions. I’d also like you to explore full AI multi-screen compositing so key moments can appear side-by-side or in picture-in-picture layouts where that helps the message land. The finished spot must communicate in a clear, formal style that’s suitable for a brand-safe environment while still feeling contemporary. I’m comfortable with whatever toolchain you prefer—Runway, Descript, Adobe Premiere Pro with Sensei,...
I have a kaggle dataset containing colored images and thermal image .. do feature extraction and then combine them and the do feature extraction on it
...drawings with dimensions, schedules, material specs and notes Acceptance criteria 1. All documents delivered in both PDF and the original editable format (DWG or Revit preferred). 2. Models must be geographically true and suitable for BIM coordination. 3. Renderings should reflect final material choices, lighting and furnishings. 4. Each discipline’s drawings are to be cross-checked for clash detection before submission. Please indicate the software you plan to use (Revit, AutoCAD, SketchUp, Lumion, etc.) and your estimated timeline for each milestone so I can align engineering reviews accordingly....
Project Name: PayTrack Industry: FinTech / Personal Finance Platform: React Nativ...Overview PayTrack is a secure mobile application that helps users track expenses, manage budgets, and analyze their financial behavior in real time. n Authentication & Security • Secure login with biometrics and PIN • Encrypted local storage • Role-based access for business accounts n Expense & Budget Management • Manual and automatic expense tracking • Category-based budgeting • Recurring expense detection n Financial Analytics • Monthly and yearly reports • Interactive charts • Exportable summaries n Main Project Keywords • Expense Tracking • Financial Analytics • Offline-First Storage • Secure Authentication • Data Visua...
...automation inside and out - Have thought about anti-detection techniques for bots How we'll work: This is a milestone-based project with 4 clear phases over about a month. I'm open to fast, agile development and I'm totally fine with you using AI assistants like Claude or Windsurf to speed things up (I use them too!). Milestones & Budget: 1. Week 1: Screen Capture & Basic Detection - $100 - Set up screen capturing for macOS - Detect the banker character with 95% accuracy - Create basic human-like mouse movement system 2. Week 2: Full Firemaking Cycle - $150 - Complete bank interaction sequence - Detect and withdraw logs - Implement fire lighting - Handle inventory management 3. Week 3: Robustness & Anti-Detection -...
I need a production-ready object detection model built, trained, and packaged so it runs smoothly on iOS and Android devices, in a modern web browser, and as a lightweight desktop application. The same model should power every platform to keep accuracy and behaviour consistent. You are free to choose the framework you are most comfortable with—TensorFlow, PyTorch, YOLOv8, Detectron2, or another proven library—as long as the final artefacts meet these requirements: • Mobile: optimised builds (e.g. TensorFlow Lite, Core ML, or ONNX) that hit realtime speeds on mid-range phones. • Web: WebAssembly/WebGL or implementation that loads in under three seconds on a standard connection. • Desktop: a small executable or Python app with GPU support when availab...
I am embarking on a full-stack build of a humanoid companion robot whose number-one mission is caring for an infant from birth through age five. Beyond the usual chore-support and graceful household navigation, the robot must excel at three core caregiving abilities: • Feeding – from bottle to basic solids, with portion control, temperature checks, and spill detection. • Diaper changing – autonomous removal, cleaning, and secure re-diapering, tracked in a hygiene log. • Health monitoring – continuous vitals, posture, and environmental safety checks, with instant mobile alerts. Rich, natural interaction is essential. The robot should talk and sing lullabies, entertain with simple games, and read stories or show age-appropriate educational ...
READ THE BRIEF BEFORE SUBMITTING ANY APPLICATION - IF YOU DO NOT, IT WILL BE REJECTED IMMEDIATELY. I WILL BE RUNNING ALL SUBMISSION AT FINAL STAGE DECISIONS THROUGH AI DETECTION AFTER HAVING RECEIVED 6 SEPARATE APPLICATIONS THAT WERE IDENTICAL. Freelancer CAD Project – Sacred Balinese Pendant (CAD Clean-Up + Final Production Exports) High Detail Jewellery CAD | Paid Work Hi designers I’m looking for an experienced Jewellery CAD specialist to help finalise a premium sacred Balinese-inspired pendant, designed for 22k gold casting (lost wax method). The pendant is already mostly complete in CAD, so this project is largely production clean-up and export preparation, not a full redesign. This is a beautiful, high-impact piece and I’m excited to bring it into...
I need an experienced electronics/PCB engineer...to ALL lamp outputs • Must drop on: - ESP32 crash - Watchdog timeout - Power instability • Safe power-up and fail behavior is mandatory fuse. WATCHDOG & RESET: • External hardware watchdog + reset supervisor • ESP32 provides heartbeat signal • Watchdog disables outputs and drops master relay on failure • Do not rely only on ESP32 internal watchdog LAMP FAILURE DETECTION: • Detect CURRENT flow • Group-level detection is acceptable POWER: • Input: 12V DC PCB & SAFETY: • Proper 220VAC clearance/creepage • Clear separation of HV and LV sections WHO SHOULD APPLY: • Experience with ESP32 hardware • Experience with AC mains / industrial elec...
...Automation Requirements 7.1 AI-Driven Content Production • AI-assisted short-form video creation (Reels, TikTok, Shorts) • AI copywriting and caption generation • Automated localization (English, French, local slang variants) • High-volume creative A/B testing Expected output: Minimum 50–100 content assets per week with minimal marginal cost. 7.2 AI-Powered Viral Engineering • AI-based trend detection and social listening • Same-day reaction content for trending topics • Meme-based and story-driven virality • Optimization of emotional triggers (confessions, debates, polls) ________________________________________ 7.3 AI-Optimized Paid Acquisition • Automated bidding and budget allocation • Creative rotation and suppression o...
...RO, BG, RU, ES Sicherheits- & DevOps-Setup: DDoS, WAF, Key Management, Logging, CI/CD Funktionen / Module: User & Creator Profiles: Galerien, Texte, Shops Token-System: Kauf, Bonus, Auszahlung, Plattformkommission Payment Gateways: Multi-PSP, Betrugs- & Chargeback-Management, KYC/AML integriert Moderation: Drei-Schichten-Modell (KI → Community → Profis) KI & Recommendation: NSFW-Detection, Abuse-Detection, Content Matching Frontend-Komponenten: Widgets für Token-Kauf, Altersprüfung, DRM, Lovense Control Projektmanagement & Dokumentation: Storyboards, technische Dokumentation, Deployment Guide Technologien & Infrastruktur: Backend: Node.js / Python / Go Frontend: React / Datenbanken: PostgreSQL, MongoDB, Redis, S3/IPFS...
I'm looking for a simple laser tag prototype using ESP32 and IR technology. This is for concept testing, not a final product. Essential features: - Scorekeeping - Player detection - Team identification Additional feedback mechanisms would be a plus but are not mandatory. Ideal skills and experience: - Experience with ESP32 and IR communication - Background in game mechanics for scorekeeping and player detection - Ability to develop a user-friendly prototype Please provide samples of similar work. Also looking for someone working out of India so that I can get my hands on the final prototype, no matter how simple it is.
...instantly whenever I paste new daily data. The layout I have in mind is simple: source data on one tab, clean summary on another. Formulas are fine; a small VBA routine or Power Query step is welcome if it keeps the workbook fast and eliminates manual intervention. Deliverables • A working .xlsx file with clearly labelled input and output areas • Built-in calculations for: 1. ATM strike detection 2. Straddle value (call + put at the ATM strike) 3. Ratio of straddle value to futures price • Brief instructions so I can drop fresh data in and see the numbers update Accuracy and transparency are critical, so please keep any helper columns visible and comment complex formulas where needed. Attached files have the format I am looking for...
...hours) with a natural ramp-up, peak, and decay pattern. - Behavioral Simulation: Implement a "Session Keeper" or interaction layer (Overlay/Iframe) to ensure real users stay on the page for 30s+ to avoid high bounce rates and trigger "Engaged Session" signals in GA4. - Stealth & Fingerprinting: For the bot-driven portion of the traffic, ensure 100% bypass of TLS fingerprinting and WebGL/Canvas detection. - Configurability: A simple dashboard or Google Sheets integration to set: Target URL, Geo-location, Device Mix (Mobile/Desktop), and Scheduled Burst times. Requirements: - Expertise in Playwright, Puppeteer, or Node.js for web automation. - Deep understanding of HTTP Headers, TLS Handshakes, and Browser Fingerprinting. - Experience with Residential/Mobil...
...CMS / Sanity Vercel, Render, Railway 3. Core Platform Features 3.1 User Dashboard ● Live MetaApi trading data ● Balance, Equity, Drawdown ● Profit Target Tracking ● Trade History (Charts + Tables) ● Monthly P&L Calendar ● Account Status (Active / Breached / Passed) 3.2 Challenge System ● Multi-phase challenges ● Configurable account sizes ● Automated rule validation ● Auto breach detection ● Trade blocking on violation 3.3 Admin Panel ● User monitoring ● Rule configuration ● Challenge management ● Manual payout approvals ● Account override controls 4. Rule Enforcement Engine All trading rules are enforced in real-time. Breaches result in automatic account suspension and dashboard status updates. Includes: ● Daily drawdown ● Maximum loss ● Minimum t...
...recommendation—Norton, McAfee, Kaspersky, or another best-in-class option) and continues through full installation, optimisation, and hardening. Key protection goals • Online banking: isolate browser sessions, enable real-time phishing & key-logger defences, and verify safe-transaction modules are active. • Confidential files: enforce on-access scanning, tune heuristic settings for maximum detection, and lock down local folders with password-protected vaults or encryption available within the AV. • Google Drive sync: ensure cloud-storage scanning is enabled, block malicious file sync, and set custom rules for shared links. What I expect when we finish 1. Antivirus installed and licensed on one Windows PC. 2. All configuration screens reviewed ...
I run several e-commerce sites and currently rely on ChatGPT, Gemini and simil...return refined versions that: • read as if written by a knowledgeable human in a consistent, professional tone • include headings, bullet points or short paragraphs where appropriate for clarity and conversion • incorporate basic on-page SEO craft (engaging title, meta description, natural keyword use) and suggest keywords when relevant, because I don’t have a preset list • pass common AI-detection checks with a score under 10 % Turnaround is next-day on weekdays so the catalog keeps moving. If the first batch meets quality and speed targets, this will become a steady, long-term stream of work. Please send a short sample rewrite of any AI paragraph you choose so I can...
...storage of raw email bodies SCOPE OF WORK (FOR DEVELOPER) Phase 1 – Core System (MVP) Email ingest (BCC + direct commands) Command parsing Daily email generation Weekly summary Priority system Draft email generation Style/cadence learning Phase 2 – Intelligence & Sales Layer Contact rotation logic Quote aging detection (email-based) Re-engagement analysis Highlight reels Monthly/quarterly outputs Phase 3 – Refinement Better cadence modelling Resource influence weighting Risk language detection Reporting polish ACCEPTANCE CRITERIA The project is successful if: I can run my day from one daily email I can manage sales without opening a CRM Drafts sound like me, but better Nothing auto-sends externally The system remembers priorities and goals Weekly ...
...full interaction—audio plus structured summary—into Zoho CRM under the correct contact or ticket. • At hang-up, the caller receives a follow-up email (template supplied) triggered from Zoho. Must-haves • End-to-end voice model only—no separate speech-to-text or text-to-speech layers. • Latency comparable to human conversation. • Natural turn-taking, interruption handling, and sentiment detection to decide whether to escalate. • Dockerised deployment with environment variables for RingCentral SIP creds, Zoho OAuth, and knowledge-base endpoints. • Clear documentation and a short demo video showing a live call routed through RingCentral into the agent and the resulting Zoho log. Acceptance criteria 1. A test call lasting at...
...via webhook or callback Bank Transfer Manual payment system: 1. Checkout option: Bank Transfer 2. Unique payment reference per order 3. Receipt upload form (jpg/png/pdf) 4. Store upload + audit trail 5. Admin panel: o Approve payment o Reject payment o Change order status 6. Email/WhatsApp notification when receipt uploaded. Fraud & safety checks • Amount mismatch detection • Currency mismatch detection • Duplicate receipt detection • Payment expiry timeout Acceptance criteria • Bank transfer order → pending → admin approves → PAID. • Receipt upload works and stored securely. ________________________________________ Milestone 3 — Order State Machine + Notifications Goal: Make payments production-ready. Deliverab...
...a cloud-based Excel file. Rules The plate image: ❌ Must NOT be stored in machine photo folders ❌ Must NOT be renamed ❌ Must NOT create folders The identification plate is the single source of truth for machine data. FLOW 3 — DOCUMENTS (DELIVERY NOTES / CMR / TRANSPORT) Input Photos or PDFs sent via WhatsApp of: Delivery notes CMR documents Transport / logistics documents Automatic Detection The system must identify document images based on: Keywords (“DELIVERY NOTE”, “CMR”, “Consignee”, “Sender”, etc.) Full-page document layout Text-heavy structured format Cloud Storage Documents must be stored separately from machine photos. Base Structure /DOCUMENTS/ ├── DELIVERY_NOTES/ └── CMR/ Optional (if date detec...
...leaderboard is loaded Handling asynchronous data sources Choosing between multiple conflicting data representations This requires a hybrid scraping approach, not a static script. Current State of the System We already have a working base scraper that was partially vibe-coded. It performs well on many sites, but it is not yet robust enough at large scale. Current limitations include: Inconsistent detection of leaderboard buttons and tabs Partial leaderboards being captured instead of full datasets Site-specific edge cases breaking otherwise valid logic Some extraction methods working well in isolation but failing in combination Your role is to improve reliability, coverage, and correctness, not to rebuild everything. Technologies & Tools in Use These are important to note befo...
...“What insurance does machine 1696 have?” The AI: Searches relevant Excel files Extracts the required data Responds in natural language No file opening required. 11. Company Phones and Contacts From WhatsApp, users can ask: “Company phone numbers” “I want to call insurance” The system returns: Correct person Role Direct phone number 12. Languages and Translation Automatic language detection Responses in the same language as the user Audio and video Automatic transcription Translation Response in configured language 13. Document Delivery Documents are delivered: As a PDF attachment via WhatsApp, or As a secure, time-limited link Optionally: Summary Relevant page Key extracted data 14. Development Phases Phase 1 &ndash...
The goal is straightforward: as soon as an ESP device is present—whether connected over Wi-Fi or broadcasting in monitor mode could be spotted in hidden places
...the Azure development. The core requirement is an unsupervised machine-learning pipeline—deployed and managed in Azure—that can ingest raw data, uncover patterns autonomously, and feed those insights into the application’s decision engine. Here is what I have in mind: data lands in an Azure Data Lake, is processed through Azure Databricks or Synapse, and the unsupervised models (clustering and anomaly-surfacing techniques are likely candidates) are trained and versioned in Azure Machine Learning. Model output then needs to be exposed via a RESTful endpoint, secured with Azure AD, and consumed by the front-end service. Everything must be CI/CD-enabled—GitHub Actions or Azure DevOps—so new experiments can move smoothly to production. Deliverables &...
...Detailed project information will be shared only after shortlisting and NDA signing. Engagement Details Type: Freelance / Contract Duration: ~3–4 months (extension possible) Commitment: Full-time preferred (role dependent) Work mode: Remote (India preferred) Start: Immediate / near-term Open Roles 1. Computer Vision / Machine Learning Engineer What we’re looking for: Hands-on experience with object detection / visual recognition Experience training and tuning ML models on real-world data Strong Python skills Comfortable working with noisy, imperfect datasets Good to have (not mandatory): Experience with aerial imagery, robotics, or field-deployed AI Some exposure to edge inference or model optimization 2. Edge / Embedded AI Engineer What we’re looking for: Exper...
...Go) AI/LLM API Integration (Critical Component): Integrate an AI API (Claude API, OpenAI, or similar) as a strategic advisor layer in the decision pipeline The LLM should receive structured game state data (hand, board, pot odds, position, opponent stats, tournament stage) and return strategic recommendations Use the LLM to handle nuanced, non-formulaic decisions: multi-street planning, bluff detection, ICM-aware play in tournament bubbles, and adapting to unusual opponent patterns Encode our expert player's knowledge into a rich system prompt / fine-tuned model that reasons like our expert would The LLM acts as the "exploitative brain" while the GTO solver provides the mathematical baseline — the system blends both outputs Implement smart caching and pre-co...
I need a production-ready object detection model built, trained, and packaged so it runs smoothly on iOS and Android devices, in a modern web browser, and as a lightweight desktop application. The same model should power every platform to keep accuracy and behaviour consistent. You are free to choose the framework you are most comfortable with—TensorFlow, PyTorch, YOLOv8, Detectron2, or another proven library—as long as the final artefacts meet these requirements: • Mobile: optimised builds (e.g. TensorFlow Lite, Core ML, or ONNX) that hit realtime speeds on mid-range phones. • Web: WebAssembly/WebGL or implementation that loads in under three seconds on a standard connection. • Desktop: a small executable or Python app with GPU support when availab...
... Send & Receive Money: Users can transfer funds internationally via bank deposit, mobile wallet, or cash pickup. Currency Conversion: Real-time exchange rates and transparent fees. Payment Methods: Support for bank cards, bank transfers, and digital wallets. Transaction Tracking: Notifications, transfer status updates, and history logs. Security & Compliance: End-to-end encryption, fraud detection, and compliance with AML/KYC regulations. Admin Dashboard: Manage users, transactions, rates, and reports. Technical Requirements: Mobile apps: iOS & Android Web dashboard for admin & users Scalable backend to handle high transaction volumes Integration with payment processors, banks, and currency exchange APIs Deliverables: Fully functional apps and web dash...
...open to your guidance on which stack best balances accuracy, cost, and future scalability, provided everything runs on a maintainable cloud or container platform. Core deliverables • Architecture and tooling proposal with rationale • Data-ingestion pipelines that screen-scrape at scale and connect to our databases • Machine-learning layer that turns raw inputs into trends, forecasts, and anomaly flags • Automated report generator (dashboards, PDFs, or both) running on a schedule • Deployment scripts and hand-off documentation Acceptance criteria 1. Daily analytics complete in under 30 minutes without manual intervention. 2. Forecast error ≤5 % against historical back-tests. 3. Codebase fully containerised or otherwise reproducible on ...
