AI-Driven Customer Reply Automation

$250-750 USD

Posted 1 minute ago

•

Ends in 6 days

$250-750 USD

Paid on delivery

I need an n8n workflow that receives incoming customer messages, sends them to either Anthropic-Claude or Google-Gemini for analysis, and returns a polished, on-brand reply without my manual involvement. The flow should handle authentication, pass a tone-of-voice prompt I will supply, and then push the generated answer back through the same channel so the customer experiences a seamless conversation. Please wire the solution so that: • n8n remains the single orchestration layer. • Model keys for Claude and Gemini live in environment variables. • I can switch between models by changing one setting, not the entire flow. I’ll provide sample conversations, response style guidelines, and access credentials for testing. The project is complete once you deliver the n8n workflow JSON, brief setup notes, and a short Loom or text walkthrough showing the automation responding correctly to at least five real messages from my test inbox.

Project ID: 40826955

About the project

40 proposals

Open for bidding

Remote project

Active 9 secs ago

Place your bid

Bid amount Email address

Benefits of bidding on Freelancer

Set your budget and timeframe

It's free to sign up and bid on jobs

40 freelancers are bidding on average $371 USD for this job

User Avatar
Flag of MEXICO

Hi, I'm Fernando, a developer from Mexico. The important part here is making the automation respond naturally while keeping the workflow simple to maintain. You shouldn't need separate workflows or duplicated nodes every time you switch between Claude and Gemini. I'd build this in n8n as a single orchestration flow, with a configurable model selection, environment-based API credentials, and your tone-of-voice instructions applied consistently to both models. The workflow will receive incoming messages, authenticate with the relevant channel, process the conversation context, generate the reply, and send it back through the same channel. I'll also include error handling and execution logs so failed requests don't go unnoticed. Before delivery, I'll test at least five real messages from your test inbox and provide the workflow JSON, setup instructions, and a walkthrough. Which customer messaging platform are you currently using? I'd like to confirm its API capabilities before choosing the integration approach. Best regards, Fernando

User Avatar
Flag of SOUTH AFRICA

Hi there, we might be new to the platform, but I'd love to share some relevant examples. Your project on creating an n8n workflow sounds right up our alley, and we would love to exceed your expectations. I understand that you need a streamlined solution to receive customer messages, analyze them with either Anthropic-Claude or Google-Gemini, and send polished responses automatically. This seamless flow will enhance your customer experience significantly. By utilizing environment variables for model keys, you can easily switch between Claude and Gemini without altering the entire workflow, ensuring maximum efficiency. While we might be new to Freelancer, we have over nine years of experience off-site, and I truly appreciate you taking the time to review our proposal. I would love to discuss this project more. Kind regards, Trichelle. Client name is and my name is Trich123.

User Avatar
Flag of SERBIA

Hi, I've built several n8n automation workflows that integrate LLMs into customer communication pipelines, so I understand the orchestration and model-switching requirements you're after. I'll set up your workflow to receive incoming messages, route them through Claude or Gemini based on an environment variable flag, apply your tone-of-voice prompt, and push the polished reply back through the same channel—all without manual steps in between. The flow will store both model API keys as environment variables, so switching between Claude and Gemini is a single configuration change rather than rewiring nodes. I'll test the complete loop against your sample conversations and real messages from your inbox to confirm responses land correctly before handoff. Since these workflows can fail silently if authentication or routing logic isn't bulletproof, I'll validate each step with real data and log any errors clearly so you can spot issues early. You'll receive the workflow JSON, setup notes for environment variables and credentials, and a walkthrough showing five end-to-end test runs from your inbox. I'm available now and will stay on for any adjustments or questions once it's live.

User Avatar
Flag of BANGLADESH

Automating customer replies without losing your brand voice or locking into one AI takes clean orchestration. With 9+ years of experience and 470+ completed projects, I can build this n8n workflow. Here's how I can help: - Keep n8n as your single orchestration layer with secure authentication and same-channel replies for seamless conversations. - Wire Claude and Gemini keys via environment variables with a one-setting switch and your tone-of-voice prompt. - Deliver the tested workflow JSON, setup notes, and a Loom walkthrough proving 5 real inbox replies work accurately. Your automation will run reliably from day one. Which channel hosts your test inbox, and should the AI reference prior thread history? Let's discuss the details in chat.

User Avatar
Flag of ARGENTINA

One risk in this project is that the model switch setting becomes an afterthought. I would address it first by defining a single "model" parameter at the workflow start, used in one HTTP Request node for either Claude or Gemini, so switching never requires editing the core logic. The first concrete step would be to build and test the authentication layer with your provided keys mapped to environment variables. That matters because if auth is not isolated correctly, everything downstream fails or your keys leak into the workflow JSON you need to share. I would then wire the inbound message, inject your tone-of-voice prompt, call the selected model, normalize the response, and push it back through the same channel. I will test against your five real messages and deliver the JSON, setup notes, and a short walkthrough showing the switch working for both models.

User Avatar
Flag of PAKISTAN

Hi, I understand you need an n8n workflow that receives customer messages, sends them to Anthropic-Claude or Google-Gemini, and returns a polished reply through the same channel. I will keep n8n as the main orchestration layer, store both model keys in environment variables, and add one simple setting for switching models. I will also use your tone guidelines and sample conversations, test the flow with at least five real messages, and provide the workflow JSON with clear setup notes and a short walkthrough. My experience with Automation, API Integration, Chatbot, and AI Chatbot Development fits this work well. Which customer messaging channel should n8n receive replies from and send the generated responses back through? Please start the chat and let’s discuss the message channel and testing access.

User Avatar
Flag of UKRAINE

Hello, Your key requirement is keeping n8n as the single orchestration layer while allowing Claude or Gemini to be selected through one setting. I would structure the workflow with environment-based credentials, a model-selection branch, your tone prompt, and a consistent return path to the originating channel. My background includes AI automation, LLM integrations, and third-party API workflows. I will use your sample conversations to refine the prompt behavior, test authentication and failure handling, then provide the workflow JSON, concise setup notes, and a walkthrough or written demonstration covering five real test messages. I can deliver the first working automation, validation, and documentation within 5 days. Best regards, Bohdan

User Avatar
Flag of MEXICO

I can build the n8n workflow so it handles the full loop—receiving messages, calling the model, and sending the reply back through the same channel. The one-setting model switch makes sense; I’ll use an environment variable for the model name too so you don’t have to edit nodes when swapping Claude for Gemini. From my work on AI agents and automation platforms like JustBot and Revnu, I’ve integrated LLM routing with environment-based credentials and tone-of-voice prompt injection across customer-facing flows. I’ll start by mapping your channel’s message format and reply endpoint, then build the workflow around a single webhook or trigger that passes the incoming text and tone guidelines to the model. I’ll add structured error handling for timeouts or rate limits so failed replies don’t break the conversation, then simulate five test messages and record the walkthrough. Could you tell me which channel the customer messages will come through—email, WhatsApp, web chat, or something else? That determines the first node and how replies are pushed back.

User Avatar
Flag of UNITED STATES

Hi, The thing that'll actually bite you here isn't the AI part, it's what happens when Claude or Gemini goes down or times out mid-conversation. I'll build in a fallback so a slow API response doesn't leave a customer hanging with no reply at all, that's the piece most people building this skip. For the build itself, n8n handles the whole flow: it grabs the incoming message, pulls your tone-of-voice prompt, sends it to whichever model you've picked, and pushes the reply back out the same channel. Model keys sit in environment variables, and switching between Claude and Gemini is just flipping one setting, no rebuilding anything. I've done this kind of message-in, AI-reply-out wiring before, it's pretty familiar ground. I can have this running, tested against five real messages, with setup notes and a walkthrough, in 2 days. Send over the sample conversations and style guide whenever you're ready and I'll get moving. Best, Emrah

User Avatar
Flag of GERMANY

Hola, "AI-Driven Customer Reply Automation" - necesitas respuestas automaticas coherentes con tu marca, con un flujo sencillo de mantener y cambiar de modelo. I'll configure n8n as the single orchestration layer, receiving authenticated customer messages, applying your tone-of-voice instructions and sending the generated reply back through the original channel. I'll integrate Claude and Gemini behind one model-selection setting, keep both API keys in environment variables, and handle API errors, timeouts and malformed model responses so a failed request does not silently leave a customer without an answer. I'll test the workflow against at least five messages from your test inbox and deliver the importable JSON with setup notes and a walkthrough. Do you want the model-selection setting to support an automatic fallback to the other provider if the selected model is unavailable? Quedo atento a tu respuesta. Beate

User Avatar
Flag of UKRAINE

Building AI-driven customer automation is a strong match for my full-stack development background and AI chatbot development skills. I can create an n8n workflow that keeps orchestration in one place, routes messages to Claude or Gemini through a single model setting, and returns polished replies using your tone-of-voice prompt. My experience with Node.js, Python, RESTful APIs, Docker, Git, and AWS supports the integration work required for authentication, environment-based model keys, webhook or channel communication, and maintainable workflow delivery. I also work with AI content creation, allowing me to focus on responses that are consistent, natural, and aligned with your brand rather than merely connecting APIs. I will deliver the n8n workflow JSON, concise setup notes, and a short walkthrough demonstrating successful replies to five test messages. Choose me to build a clean, flexible automation you can operate without manual intervention.

User Avatar
Flag of VIETNAM

$=== SUPER MATCH ===$ I can build this as a modular n8n workflow with one configurable model selector for Claude or Gemini, keeping both API keys in environment variables. Incoming messages will pass through authentication, your tone-of-voice instructions, and context-aware prompt handling before the polished reply is returned through the original channel. I’ll include error handling, model-switching logic, an importable workflow JSON, setup notes, and a walkthrough. I’ll also test at least five real messages from your test inbox and verify the replies reach the correct channel. Which messaging platform or inbox will trigger the workflow? Looking forward to working with you.

User Avatar
Flag of VIETNAM

⭐⭐⭐⭐⭐ Hi there, Can you confirm you need one n8n workflow that receives messages, selects Claude or Gemini through one setting, and sends the polished reply back through the same channel? The key priorities are reliable authentication, secure environment-variable keys, consistent tone control, and seamless replies without manual intervention. The workflow must also be easy to test and switch between models. My background includes AI product development and Python-based automation, including OpenAI, Azure, AWS, and TensorFlow solutions. I can apply that workflow discipline here while learning the required n8n and model-specific setup rather than overstating prior experience. ❓ Which customer messaging channel and inbox should the workflow connect to? ❓ Should the model setting be changed through an n8n variable, credential, or configuration node? I’ll focus on clear orchestration, secure key handling, reusable model selection, and a concise setup/text walkthrough covering five test messages. I’m ready to begin immediately. I look forward to your reply and discussing the implementation. Best regards, Phuong Nam D

User Avatar
Flag of BANGLADESH

Hi, You need incoming customer messages answered automatically in your brand voice, with the reply sent back through the same channel. I'll build the n8n workflow with Claude and Gemini switchable through one setting and keys stored in environment variables. With 7+ years in web development, I'll use n8n, API integration, and AI model connections to build an automation that keeps conversations seamless without manual work. Which channel will the messages come from, such as your website chat, e ma il, or a messaging platform? Regards, Sakline P.S. Sharing your tone-of-voice guidelines and a few sample conversations will help me tune the first test run quickly; I'll also include setup notes for switching models.

User Avatar
Flag of PORTUGAL

Hello, I’ll build this n8n workflow with a modular design where the LLM selection is controlled by a single variable, so switching between Claude and Gemini only requires updating one setting. I recently created a similar flow for a client where the API keys were stored in environment variables and the workflow handled retries for failed requests, reducing their error rate by 30%. The setup will include a pre-processing step to inject your tone-of-voice prompt before sending messages to the model, ensuring consistency without manual oversight. This automation will cut response times by 90% while keeping the conversation natural, since the workflow will validate outputs against your style guidelines before sending replies. After delivery, I’ll include a 14-day support period to fix any edge cases that come up during testing. If this approach fits your needs, I’m happy to review your sample conversations and credentials to tailor the workflow. Thanks, Marcio.

User Avatar
Flag of UNITED STATES

Greetings, I'd build this around one goal: every incoming customer message receives a consistent, professional response without requiring your intervention. Using n8n as the central orchestration layer, I'll connect your messaging channel to a shared AI response engine, incorporate your sample conversations and brand guidelines, and configure Claude/Gemini selection through one editable setting. I'll also add safeguards for failed API requests and prevent duplicate responses. You'll receive the complete workflow JSON, concise setup instructions, and documented results from five real-message tests. Similar Work / Portfolio: https://www.freelancer.com/u/muhammadkhan32 Do you want the assistant to remember previous messages within each customer conversation, or respond independently to every enquiry? I'd welcome a chat about making the experience as natural and consistent as possible. Warm regards, Muhammad Khan

User Avatar
Flag of INDIA

I can build an n8n workflow that receives each incoming customer message, sends it to Claude or Gemini using the supplied tone-of-voice prompt, and posts the polished reply back to the same channel automatically. I’ll start by creating a reusable HTTP request node that reads the model key from an environment variable, then add a switch node so changing a single parameter selects Claude or Gemini. The workflow will include authentication steps, the prompt injection, and a final node that pushes the answer back, all wired as a single JSON export. I’ll test it with your sample inbox and record a short Loom walkthrough showing five real messages processed correctly. I can have it working within 2 days. Dileep

User Avatar
Flag of BANGLADESH

Hi, With 7 years of web development and AI automation experience, I can build your n8n workflow to receive customer messages, generate on-brand Claude or Gemini replies, and respond through the original channel. I'll configure environment-based API keys, one-setting model switching, authentication, and error handling. You'll receive tested JSON, setup notes, and five-message verification. Let's discuss in chat to proceed further.

User Avatar
Flag of AUSTRALIA

Hello, I’d love to help you build this AI-driven customer reply automation in n8n. I’ve worked on similar AI automation flows where incoming messages were routed through LLMs, polished with a brand-specific tone, and sent back automatically with minimal manual handling. For this project, I can set up a clean n8n workflow where the channel stays fully managed inside n8n, while Claude and Gemini are both connected through environment variables so switching between them is just one simple setting change. I’ll also make sure the tone-of-voice prompt is easy to pass in, and the reply goes back through the same customer channel for a seamless experience. I recently built a comparable automation for a support team where customer inquiries were analyzed, categorized, and answered using AI with consistent brand voice, which reduced manual back-and-forth and kept responses natural. A couple of quick questions: 1. Which customer channel should the workflow support first? 2. Do you want fallback handling if the model fails or returns a weak response? I’d be happy to walk you through the setup on a call. Best regards, Mehreen

User Avatar
Flag of GERMANY

The main difficulty is keeping model switching, credentials, tone rules, and channel replies consistent in n8n. We will build the workflow to receive each message, route it to Claude or Gemini through one setting, return an on-brand answer, and deliver the n8n JSON, setup notes, and a five-message test walkthrough. We built a gateway to multiple AI models with access control and automatic balance top-ups, so this API integration is familiar territory. Which customer inbox or messaging channel should the workflow connect to?

内容来自 Freelancer News,版权归原作者所有。

阅读原文 ↗

用 HomepageTab 新标签页关注 RSS,打开浏览器就能看到更新。

安装 HomepageTab