AI Integration Projects

opportunity

Connector between Claude and CRM from Dynamics Microsoft

Our company is seeking a software developer to build and finalise an integration connector for our Microsoft Dynamics 365 Sales (Dataverse) environment. The connector must let internal staff create business records using plain natural-language instructions in English or Spanish – for example, “create a quote for client X with 5 units of product Y” – and have the system translate that instruction into the correct Dynamics operations. A working prototype already exists (Python, plus a JavaScript data-access layer). The goal of this engagement is to harden the prototype into a reliable, maintainable, production-ready tool, validate it against our live environment, and document it for our team. This document describes the existing components, what works, what remains to be done, the required skills, and the deliverables. It is intended to give a candidate enough information to scope the effort and quote accurately.

opportunity

Looking for an Expert ChatGPT & Claude

I am looking for someone who can help me with the following: ✨ Setting up or migrating a Claude account ✨ Exploring possibilities to integrate Claude with my Instagram account ✨ Smart AI setup for content creation, workflows and working more efficiently A compensation of €100 is available for this assignment. Do you have experience with AI, automations and integrations between tools and social media, or do you know someone who does? Please send me a message or tag someone who might be a great fit.

Supabase Database Admin for Product Data

We are seeking an experienced Supabase specialist to support an existing project that manages structured product data. The role involves reviewing and understanding the current database architecture, updating records using newly provided datasets, improving data quality, and recommending database structure enhancements to better support frontend/UI functionality. The ideal candidate should be comfortable handling the full data workflow, including data cleaning, transformation, validation, and secure uploads into the existing Supabase environment. Strong documentation practices are important, as all work, changes, and validation steps should be tracked clearly through GitHub. This position also includes working closely with the project owner to identify opportunities for automating recurring data update processes and improving overall workflow efficiency. We are looking for someone who: • Has hands-on experience with Supabase and relational databases • Can work independently with minimal supervision • Maintains high attention to detail and data accuracy • Is capable of moving quickly while preserving data integrity • Understands structured validation and quality-control practices • Is comfortable documenting technical work and processes Data accuracy and validation are critical for this project, and all updates will be reviewed carefully throughout the engagement.

I need an urgent make.com scenario debugged.

Urgent Make.com scenario debugging required: my automation recently failed and needs immediate diagnosis and repair. Seek an experienced Make.com integrator to investigate scenario modules, traces and execution history, identify root cause(s), implement robust fixes, optimize error handling and suggest preventive measures. Deliver clear summary of findings, applied changes and recommendations for resilience. Timely completion and demonstrable troubleshooting expertise are essential.

opportunity

I want rebuild our compay website with AI

I want ai developer who can rebuild our company website as casual with ai If you are confident and have rich experience in this field, please apply on this job Thanks

Intergration PB WB

Project title: Pipedrive, QuickBooks and Mintsoft API automation for 3PL order workflow We are looking for an experienced API/automation developer to help build an order workflow between Pipedrive, QuickBooks Online and Mintsoft. Current systems: - Pipedrive is used by the customer’s sales team - QuickBooks Online is used for invoicing/accounts - Mintsoft is used by our 3PL business for order fulfilment, pick/pack and dispatch Required workflow: When a deal in Pipedrive is moved to a specific stage, for example “Ready for Fulfilment”, the automation should: 1. Pull the deal, organisation, contact and product line information from Pipedrive. 2. Validate that required order fields are complete. 3. Check/match customer and product/SKU data. 4. Create the fulfilment order in Mintsoft. 5. Create or prepare the invoice in QuickBooks Online. 6. Update the Pipedrive deal with the Mintsoft order number, QuickBooks invoice number and automation status. 7. Log errors clearly if required fields, SKUs, customer details or API calls fail. The solution can be built using Make.com, Zapier, n8n, Pipedream or a lightweight custom API/middleware, but we are looking for the most reliable and cost-effective option. Important: - Experience with Pipedrive API/webhooks is required. - Experience with QuickBooks Online API is required. - REST API and JSON experience is required. - Experience with Mintsoft API or warehouse/order management APIs would be a strong advantage. - We need proper testing and a simple handover document. Please respond with: 1. Similar integrations you have completed. 2. Your recommended approach. 3. Fixed price for phase 1. 4. Estimated timescale. 5. What access/information you would need from us.

Develop Microsoft Power Automate Flow + PDF & Excel integration

I am looking for a developer to write two Power Automate Flows to do the following When an email containing a purchase order arrives at an inbox, to examine the PDF attachment and extract the order number item code, description and quantity. This information is to be added to an excel spreadsheet table. I then need a macro on the Excel spreadsheet with a button, so that I can select the order number, then print the correct quantity of labels with the item and description. So the work comprises Developing PA flow to extract information from PDF and add to excel table Designing Excel macro to print labels off once items have been added to the table. I have Power Automate subscriptions and sharepoint folders etc. A sample purchase order is attached - all purchase orders will be of this format

opportunity

AI-Powered Price Scraper & Monitoring System (Multi-Website)

We are looking for an experienced developer to build a scalable AI-powered price scraping and monitoring system. The system should automatically extract product pricing data from multiple e-commerce websites and store it in a structured database for monitoring and analysis. The system must support multi-tenant architecture, role-based permissions, subscription tiers, and Stripe payment integration. The goal is to allow different companies to monitor product prices across multiple websites, with usage limits based on subscription plans. Project Scope 1. Target Websites • Scrape product prices from 7–10 e-commerce websites • Support dynamic content (JavaScript-rendered pages) • Proxy rotation & anti-bot handling • Scheduled scraping • Historical price tracking • Price change alerts (email or webhook) • Handle pagination and product variations 2. Multi-Tenant Architecture • Super Admin role • Manage all companies • Manage subscription plans • View system-wide usage • Suspend / activate companies 2. Data Extraction • Extract product name • Current price • Original price (if available) • SKU / Product ID • Availability status • Timestamp 3.1 Company Admin role • Manage company users • Set scraping targets (websites & products) • View company usage stats 3.2 Company Users • View price tracking dashboard • Access only assigned websites/products 3.3 Subscription & Usage Limits System must support different plan levels: Each plan should control: • Maximum number of websites • Maximum number of products • Scraping frequency (e.g., 1h / 3h / 6h / 24h) • Maximum concurrent scraping jobs • Historical data retention length Stripe Integration • Stripe subscription integration • Monthly / Yearly billing (7 days free trial) • Webhook handling for subscription status updates • Automatic feature unlock based on plan • Auto suspend account if payment fails • Admin ability to manually upgrade/downgrade plan 4. AI-Assisted Selector Detection • Use AI or intelligent selector logic to detect price elements • System should adapt if minor DOM changes occur • Minimize manual reconfiguration 5. Infrastructure • Proxy rotation support • Anti-bot handling • Headless browser support (e.g., Puppeteer / Playwright) • Scalable deployment (Docker preferred) 6. Database & Storage • Store data in MySQL • Historical price tracking • Ability to compare price changes 7. Monitoring & Automation • Scheduled scraping (e.g., every 1–6 hours) • Email or webhook alerts when price changes • Logging and error reporting 8. Dashboard • Admin and users dashboard • Search by product • View historical price chart Technical Requirements Preferred stack: • Laravel • Playwright / Puppeteer / Scrapy • REST API architecture • Docker deployment Deliverables • Fully working scraping system • Deployment guide • Source code • Documentation • 2 weeks post-delivery support Bonus Experience with anti-bot bypass, rotating residential proxies, and large-scale scraping is highly preferred. If interested, please include your portfolio and examples of similar scraping projects.

opportunity

Out‑of‑Hours WhatsApp Business Automation

Out-of-Hours WhatsApp Business Automation – Project Brief Project Overview I am seeking an experienced developer to integrate a WhatsApp Business automation system for my business. The purpose of this project is to design and implement the core framework and logic for handling WhatsApp enquiries out of hours (8:00pm–8:00am UK time), when messages are currently unmanaged. Due to the nature of the business, enquiries are often time-sensitive. If customers do not receive an immediate response, they frequently look elsewhere. The system must therefore acknowledge and guide enquiries instantly. This phase is focused on system architecture, integration, and scalability, not on finalising message wording or content. Project Objectives The system must: - Operate only between 8pm and 8am (UK time) But have ability to extend these hours for Bank Holidays etc - Instantly respond to incoming WhatsApp messages during those hours - Distinguish between new customers and existing customers - Use rules-based logic and structured conversational flows - Provide website links so customers can book services immediately - Allow seamless handover to a human when required Important Scope Clarification This is not an AI or GPT-style chatbot. The solution should be predictable, controlled, and fully rules-based. I will be adding and refining large amounts of information later, including FAQs, pricing guidance, and service descriptions. The system must support extensive pre-programmed content that can be easily expanded over time. Functional Requirements Operating Hours Logic: - Automation active only between 8:00pm and 8:00am - Outside of these hours, WhatsApp should revert to manual handling Customer Identification: The system must identify new customers and existing customers using a method proposed by the developer. Conversational Framework: - Modular conversation flows - Keyword or menu-based routing - Support for multiple pre-programmed response paths Booking & Website Integration: - Ability to insert booking and information links into automated replies Human Handover: - Automation must stop immediately once a human responds Data Capture & Logging: - WhatsApp number - Date and time - New vs existing customer flag Scope of This Phase Included: - WhatsApp Business API integration - Out-of-hours automation logic - New vs existing customer handling - Scalable conversational framework - Basic documentation Excluded: - Final message wording - Full FAQ or pricing setup - Advanced AI or NLP systems Technology Open to recommendations, but likely to involve WhatsApp Business API, a backend service, and basic data storage. What to Include in Your Proposal Please include relevant experience, recommended platform, approach, timelines, costs, and any ongoing fees.

Require the implementation of automation in our service delivery

We use M365 and are looking for a developer to assist us with implementing AI as a business solution for processes within our business. This is typically: 1. Extracting attachments and performing risk assessments using CoPilot. 2. Information from MS Lists triggers workflows 3. Using PowerBI to report on service delivery 4. Effective design of Sharepoint to use as a tool for sharing information with other professionals outside the organisation.

opportunity

Integrate Anthropic to remove manual labour

Hi all, Internally we have some manual processes where we want to integrate Ai into the business. The main process that we require would be below: Sales Process Web enquiries come in and we send a quote. We don’t have fixed prices as these can change every quarter but we would like the ability to use Ai to implement this job. For regular clients that email in we would also have to quote, some product is standard and straight off the price list which I assume AI could help. Bespoke would still have to be done manually. Order processing We get email orders from clients. They send masters and artwork via email. We upload the content to a portal and the artwork is checked by us through a third-party programme. This is all a manual process. We send the artwork by PDF email to the client for approval. Invoicing Supplier sends an XML file every day which we upload onto our bespoke system which then takes the data and creates our invoice using the data and we add in our pricing. Paper waste With every order there are a lot of emails going back and forth with many changes on the order along the way. We print these emails out and collate this with the invoice as these would include some extra charges. We invoice the extra charges and email to clients. The paperwork is then scanned and stored on our servers and once this is done the actual paper is then disposed of. i would like to automate these processes using Anthropic and i'm looking for someone to help me do this.

opportunity

AI Coordinator

AI Operations Expert for Multi-Channel Ecommerce Overview I run a fast-growing online brand selling through multiple storefronts. I need help from an AI operations expert to streamline our marketing and operations across all platforms. Scope of work - Clone and redeploy marketing assets across Amazon, Etsy, TikTok Shop, website, eBay, Facebook, and Instagram. - Design automated funnels turning social followers into customers using chatbots and dynamic offers. - Monitor and optimize AI stack, including data pipelines, prompt tweaking, insights surfacing, and documentation. - Ensure platforms pass brand-review spot checks and deploy unified creative assets. - Develop automated funnels ready for attribution from first social touch to checkout. - Create a clear dashboard or weekly report with performance metrics and system alerts. - Document step-by-step SOPs for junior team members to run basics. References - Current tools: Shopify, Zapier, Klaviyo, Meta Ads Manager, Amazon Ads, ChatGPT; Open to Make, Tray.io, Manychat. Additional information I'd love to hear how you can prove a more efficient mix of tools. Let's set up a call and map the roadmap together. Company details - I run a fast-growing online brand with storefronts on Amazon, Etsy, TikTok Shop, website, eBay, Facebook, and Instagram. - Ecommerce Automation Type Data syncing, Marketing campaign, CRM workflows, Task automation, Email automation ntegration Points Website, Shopify, Social Media, Email Platforms , , , Websites, , , Instagram, Shopify, Zapier Ideal candidate (Optional) Budget and timeline (Mandatory)

Make a (production/demo) level install of a DGX Spark cluster

Hi, I'm looking for an experienced AI infrastructure specialist — or a passionate enthusiast with solid hands-on experience — to build a robust, flexible, and high-performance backend foundation for OpenClaw (an open-source autonomous AI agent) on two NVIDIA DGX Spark systems connected via 200 GbE (ConnectX-7). The goal is a stable, demo-ready environment that showcases the power of open-source models. OpenClaw will run on a separate system and connect easily via OpenAI-compatible APIs (or equivalent best-practice interfaces). Everything should prioritize: Easy remote access and external connectivity (via Tailscale/ZeroTier). Fast performance within the hardware's unified memory constraints. Simple model switching/adding later (hot-swappable where possible). Persistent services with web UIs for live demos. On-demand tools that can spin up/down cleanly. Hardware & Current State 2× NVIDIA DGX Spark (Grace Blackwell, 128 GB unified LPDDR5x memory each, ARM64). 200 GbE interconnect + 1 GbE internet links. I have a basic cluster setup, timeshift snapshots, ZeroTier, and Tailscale already running. You're welcome to rebuild from scratch if that's cleaner and faster. Core Requirements (Persistent Where Possible) vLLM as primary inference engine with a large-context main model (e.g., Nemotron 120B or equivalent). Must support easy switching to newer models. Whisper (or best-practice alternative like faster-whisper) – ready for OpenClaw API integration. Piper TTS – ready for OpenClaw API/text-to-voice integration. All persistent services should run with clean web UIs for demo purposes. On-Demand Tools (Configured for Easy External/Tailscale Access + Web UIs) Ollama + web UI (for specific or scheduled models). OCR model + workflow (let's discuss the best option—e.g., EasyOCR/PaddleOCR—and data saving/integration with other tools). Image generation (primary for OpenClaw use) with multiple models available. LoRA training tools for image generation. RAG / vector DB (choose the best integration with OpenClaw and other tools—e.g., Qdrant, Chroma, or Milvus). Multi-agent capable dev tools / environment. Central Portal & Usability A single web-based portal (e.g., OpenWebUI or equivalent) for central access to all tools, easy model switching, admin controls, and live demos. Nice-to-Have / Optional Enhancements (quote separately if interested) Full 2-node clustering with tensor parallelism (e.g., using the open vLLM-DGX-Spark repo or Ray/NCCL). Docker Compose / lightweight Kubernetes orchestration for easy updates and portability. Monitoring dashboard (Prometheus + Grafana). NVIDIA NIM microservices for optimized inference. Any other best-practice tools you recommend for integration, speed, or flexibility. Your Profile You're deeply familiar with these tools (or eager to dive in as an enthusiast), NVIDIA DGX systems (especially Spark/Grace Blackwell), multi-node inference (vLLM, tensor/pipeline parallelism), Docker/containerization, and API integrations. You understand VRAM/unified-memory optimization and can make everything work together smoothly. Bonus if you have experience with OpenClaw, OpenWebUI, RAG pipelines, or agent frameworks. Proper English communication (written and spoken) is a must for smooth collaboration. Timeline & Expectations We have a tight deadline — I need a stable, running environment live as soon as possible. You're completely free to experiment, test, and play around with different configurations during setup, but the priority is delivering a functional, demo-ready system quickly. Speed matters, while still maintaining quality and stability. Compensation Competitive hourly rate (fully flexible and based on your region, experience, and the exact scope) or a fixed-price project bid if preferred. I'm completely open to discussion—propose whatever rate works best for you and your location. This project serves as a test case for potential further collaboration and ongoing work if it goes well. If you're the right fit, there will be plenty of exciting follow-up opportunities. Work Style & Availability This is remote work. I am completely flexible on working hours and not EU-bound. As long as you're excellent at what you do, I'm happy to work with talent from anywhere in the world (including low-income countries—great people deliver great results everywhere). If this sounds like a good fit, reply with: Your relevant experience (especially with DGX Spark, vLLM multi-node, or similar stacks — enthusiasts with strong practical knowledge are very welcome). Rough timeline and cost estimate (with your proposed rate). Any questions or suggested improvements. Looking forward to building something powerful together!

Senior AI/ML Engineer

I’m upgrading our risk-and-compliance platform and need an EU-based engineer who can blend robust software craftsmanship with hands-on machine-learning expertise. The mandate is to automate decision-making across key customer-facing features by fusing classical ML techniques with Large Language Models. You’ll design the end-to-end pipeline in Python, harnessing PyTorch, TensorFlow or scikit-learn to train, evaluate and continuously improve models, then ship them as production-ready services that scale and stay maintainable. Solid experience integrating models into distributed systems, setting up meaningful evaluation metrics and communicating findings to non-technical stakeholders will be crucial, as you’ll work side-by-side with product, legal and data teams. Deliverables I’ll review: • Clean, well-documented Python codebase that builds, tests and deploys the chosen models • Containerised service or microservice exposing the decision-automation endpoint • Evaluation report covering model performance, drift-monitoring strategy and rollback plan If you have 4+ years of professional software engineering in production ML settings and can articulate your design choices clearly in English, I’d like to hear how you would tackle this challenge.

opportunity

AI-Powered Price Scraper & Monitoring System (Multi-Website)

We are looking for an experienced developer to build a scalable AI-powered price scraping and monitoring system. The system should automatically extract product pricing data from multiple e-commerce websites and store it in a structured database for monitoring and analysis. The system must support multi-tenant architecture, role-based permissions, subscription tiers, and Stripe payment integration. The goal is to allow different companies to monitor product prices across multiple websites, with usage limits based on subscription plans. Project Scope 1. Target Websites • Scrape product prices from 7–10 e-commerce websites • Support dynamic content (JavaScript-rendered pages) • Proxy rotation & anti-bot handling • Scheduled scraping • Historical price tracking • Price change alerts (email or webhook) • Handle pagination and product variations 2. Multi-Tenant Architecture • Super Admin role • Manage all companies • Manage subscription plans • View system-wide usage • Suspend / activate companies 2. Data Extraction • Extract product name • Current price • Original price (if available) • SKU / Product ID • Availability status • Timestamp 3.1 Company Admin role • Manage company users • Set scraping targets (websites & products) • View company usage stats 3.2 Company Users • View price tracking dashboard • Access only assigned websites/products 3.3 Subscription & Usage Limits System must support different plan levels: Each plan should control: • Maximum number of websites • Maximum number of products • Scraping frequency (e.g., 1h / 3h / 6h / 24h) • Maximum concurrent scraping jobs • Historical data retention length Stripe Integration • Stripe subscription integration • Monthly / Yearly billing (7 days free trial) • Webhook handling for subscription status updates • Automatic feature unlock based on plan • Auto suspend account if payment fails • Admin ability to manually upgrade/downgrade plan 4. AI-Assisted Selector Detection • Use AI or intelligent selector logic to detect price elements • System should adapt if minor DOM changes occur • Minimize manual reconfiguration 5. Infrastructure • Proxy rotation support • Anti-bot handling • Headless browser support (e.g., Puppeteer / Playwright) • Scalable deployment (Docker preferred) 6. Database & Storage • Store data in MySQL • Historical price tracking • Ability to compare price changes 7. Monitoring & Automation • Scheduled scraping (e.g., every 1–6 hours) • Email or webhook alerts when price changes • Logging and error reporting 8. Dashboard • Admin and users dashboard • Search by product • View historical price chart Technical Requirements Preferred stack: • Laravel • Playwright / Puppeteer / Scrapy • REST API architecture • Docker deployment Deliverables • Fully working scraping system • Deployment guide • Source code • Documentation • 2 weeks post-delivery support Bonus Experience with anti-bot bypass, rotating residential proxies, and large-scale scraping is highly preferred. If interested, please include your portfolio and examples of similar scraping projects.

opportunity

Make.com Automation Fixes

Make.com Automation Fixes + Google Business Profile Posting + Trustpilot Auto-Reply **Description:** I need an automation expert to fix and improve multiple Make.com scenarios. --- ## **Make Automation Issues** * **AUS scenario:** Instagram post image looks distorted [https://eu2.make.com/1290266/scenarios/7938683/edit](https://eu2.make.com/1290266/scenarios/7938683/edit) * **H2O scenario:** Facebook uses the same image for every post [https://eu2.make.com/1290266/scenarios/7874013/edit](https://eu2.make.com/1290266/scenarios/7874013/edit) * **Note (Instagram modules):** Instagram posting previously had issues (not posting) on all social media posting scenarios, but it is currently working now. Please only investigate if it breaks while you are working on the project. --- ## **Google Business Profile (Google My Business) Posting** Add **Google Business Profile posting** to these Make.com scenarios. This should post an **image + caption update**. * **AUS:** [https://eu2.make.com/1290266/scenarios/7938683/edit](https://eu2.make.com/1290266/scenarios/7938683/edit) * **H2O:** [https://eu2.make.com/1290266/scenarios/7874013/edit](https://eu2.make.com/1290266/scenarios/7874013/edit) * **Automation-Ai:** [https://eu2.make.com/1290266/scenarios/7939225/edit](https://eu2.make.com/1290266/scenarios/7939225/edit) * **Group Escape:** [https://eu2.make.com/1290266/scenarios/7665611/edit](https://eu2.make.com/1290266/scenarios/7665611/edit) --- ## **Trustpilot Auto-Reply** * Auto-reply to Trustpilot reviews is not working anymore (previously set up using UiPath). * I need this working again with a stable solution --- ## **Requirements** * Strong experience with **Make.com** * Experience with **Facebook / Instagram posting automations** * Experience with **Google Business Profile posting** * Experience with **Trustpilot automation** --- ## **Deliverables** * All issues fixed and tested * Google Business Profile posting added to all listed scenarios (image + caption) * Trustpilot review auto-reply working again --- ## **Access** I will provide access to Make.com and any required platform access once we start.