I am looking for a Make.com freelancer to create

Hello!

I have experience working with Make.com, Telegram Bot API, and integrations with AI (Claude / OpenAI) and databases (Supabase), so I can effectively set up the required scenario.

I will do:
- setting up the scenario in Make (triggers, logic, message processing)
- integration of Telegram ↔ Claude API ↔ Supabase
- saving the history of dialogues and user interests
- checking user statuses (active / stop)
… - processing READY / FLAG + integration with Google Sheets and notifications
- stable operation without failures

I will implement everything without unnecessary code — purely through Make scenarios, as needed. After completion, I will explain the logic and hand over the finished system.

I can start immediately.

Show original

Translate

Hello! The task is clear: classic architecture of an AI agent on Make.com with memory in Supabase. I have extensive experience in developing complex Telegram systems and integrating external APIs (Claude, OpenAI) with databases. I will implement the following logic:
1. Database: I will create tables in Supabase with the correct data types and indexes for fast retrieval of history.
2. Make scenario: I will set up filtering (Router) to check the status of active and aggregate context (JSON array of the last 20 messages + profiles).
3. Claude logic: I will correctly pass the system prompt and process the parsing of tags [INTRO_READY] and [FLAG] through Make tools for further recording in Google Sheets and alerting.
I will ensure stable operation and quick response of the bot (up to 1 second at the scenario logic level). Which specific Claude model do you plan to use (Haiku for speed or Sonnet for better logic)?

Show original

Translate

Роберте, вітаю!

Make.com × Claude API × Supabase × Telegram — мій щоденний стек. Зробила з десяток подібних сценаріїв за останні місяці (останній — CRM webhook → Claude класифікація → запис в Postgres з історією діалогу).

По задачі:
1. Supabase — три таблиці (users/interests/conversations) з індексами на telegram_id + created_at (швидка вибірка останніх 20). RLS політики якщо потрібно.
2. Make.com сценарій — Telegram trigger → Supabase lookup по telegram_id/status → гілка active/non-active → HTTP module до Claude API (Anthropic) з системним промптом + контекстом → Telegram reply → dual insert (user + assistant messages). Error handler на випадок overloaded/timeout.
3. Flag parsing — regex на [INTRO_READY] і [FLAG]: перший пише row у твій Google Sheet, другий відправляє notify в Telegram (твій ID).

… Working with Anthropic API з лютого — знаю rate limits, cache points, коли вмикати streaming. На Supabase SQL + REST пишу щодня.

2000 UAH, 2 дні (день на setup + пів дня на тести: флаги разом, довгий контекст, новий юзер, fallback на overloaded).

GitHub: github.com/Evge14n
35-40 год/тиждень, UTC+2, старт завтра вранці.

Дякую за розгляд!
Ольга

Good day, I am ready to solve your task, I will do it in Make or n8n, feel free to contact me. I might help reduce the cost of the project.

Show original

Translate

Hello, I confirm that I can complete the task of creating an integration in Make.com. To solve this task, I will use my knowledge of creating scenarios and connecting various services in Make.com, as well as apply my programming experience to create the necessary scripts. If there are any specific requirements for the project, please provide more detailed information. I am ready to start working and create an effective solution for your project.

Show original

Translate

Good day!

I am implementing an AI assistant through Make.com with integration of Telegram Bot API, Claude, and Supabase, including flag handling and logging in Google Sheets.

I have already set up a similar process on Make.com:
https://freelancehunt.com/showcase/work/make-com-integromat-telegram-asistent/1996765.html

I am ready to discuss the details and start working immediately.

Show original

Translate

Hello, I can implement it on make. Write to me - we will discuss in more detail!

Show original

Translate

Good day, Robert!
I have reviewed the task - the technical specifications are very specific, everything is clear. I have experience in process automation and building AI agent logic.

The stack Make.com + Supabase + Claude API is completely clear to me. I will implement correct flag checks in the database, the logic for Claude's responses, and logging of all actions. I will do everything carefully, according to your structure.

I am ready to discuss the details and start working. I am in touch!

Show original

Translate

Good day!
I have experience working with Make.com and n8n, as well as experience with Claude. I have also created bots for Telegram with AI.
I am ready to take on your project.

Show original

Translate

Hello!!!

I specialize in no-code automations through the service Make.com. You can view my profile and portfolio.

I have practical experience in implementing various types of automations and working with TG bots, different LLMs, and databases.

Feel free to reach out))

Show original

Translate

Hello! I can implement it. Write to me privately to discuss all the details. I will be glad to cooperate!

Show original

Translate

Good day!

I am ready to take on the project of setting up the AI assistant Gekko in Telegram. I have extensive experience working with Make.com, Telegram Bot API integrations, Claude API, Supabase, and Google Sheets. I will quickly and professionally set up all the described automation logic.

Message me privately to discuss the details.

Show original

Translate

Hello.

I create bots on NodeJS and through Make. I am ready to take on the task. Write to me, we will discuss.

Show original

Translate

Hello, for over 4 years, we have been a small team developing web tools and integrating AI into workflows, so we are ready to quickly and efficiently set up automation for the Gecko bot. We will build a reliable script that takes into account the verification of active status, saving the history of dialogues in Supabase, and instant response to the flags [INTRO_READY] and [FLAG] in Claude's responses. Our experience of over 4 years allows us to accomplish such a task in 1-2 days with a guarantee of correct processing of JSON data between Anthropic and your database; be sure to check our completed projects: freshagro.com.ua, farfieworldwide.com, rivnekolo.com.

Show original

Translate

Good day, my colleague and I have been professionally engaged in technical design and business process automation through low-code tools for over 4 years, so we will help you assemble a "turnkey" scenario in Make.com. We will implement a frontend audit of your request, set up the database structure in Supabase considering user connections and their interests, and also implement complex filtering logic for Claude's responses for triggers in Google Sheets. Our experience of over 4 years is confirmed by successful cases of developing AI bots with deep memory context, check our quality of execution: drkukharevich.rivne.ua, crave-agency.com.ua, jk-solution.com.ua.

Show original

Translate

Hello, my partner (designer + full-stack) and I have been developing complex automation and integration systems with LLM for over 4 years, so we will professionally set up the scenario for your assistant Gekko. We will design the architecture of tables in Supabase, ensure technically flawless logic for status checks and context retrieval from 20 messages in Make.com, as well as configure the handling of specific flags for Google Sheets and notifications. Our experience of over 4 years allows us to guarantee stable operation of the Telegram-Claude-Supabase connection without unnecessary delays, take a look at our technical works: hyperfi.tech, espressolab.com.ua, hudi.com.ua.

Show original

Translate

Hello, I am ready to complete your task, I have experience working with similar projects.

Show original

Translate

Hello! I am the founder of the engineering agency Vaysed. Your technical specification is excellent — it is a classic pattern for building an AI agent with contextual memory.

You mentioned Make.com or n8n. As engineers specializing in resilient Telegram ecosystems, we strongly recommend building this flow specifically on n8n (or as a lightweight microservice on Node.js). Make.com will burn a huge number of paid operations in each cycle (checking the user -> pulling 20 messages -> querying Claude -> recording 2 new messages), while n8n on its own server operates without limits and delays.

Here’s how we will technically implement your task "turnkey" in 1-2 days:

Database structure (Supabase): We will deploy tables for users, interests, and conversations. We will definitely set up the correct indexes for the conversations table (by telegram_id and created_at) so that fetching the last 20 messages occurs in milliseconds and does not create delays before Claude's response.

Gatekeeper and Context: We will set up a strict check (Router). If the user is inactive — the flow is instantly interrupted by sending a link to the application. If active — we aggregate data from Supabase into a single context array (system prompt + interests + formatted dialogue history) and pass it to the Anthropic API.
…
Regex parsing of flags: The response from Claude will not be blindly sent to the user. We will set up an intermediate parsing module. If the system detects the regular expression [INTRO_READY], a parallel branch of the scenario will send data to Google Sheets. If [FLAG] — you will receive an instant alert in personal messages, and the technical tag will be removed from the final message for the user.

We are building an infrastructure that works reliably and does not crash under load. You can evaluate our approach to bot development and AI integrations in our portfolio: https://vaysed.me/.

I am ready to discuss the details and start the setup immediately!

Show original

Translate

Robert, glad to have the opportunity to introduce myself, my name is Yuri) I was intrigued by your project, it's very interesting, I am ready to start this evening! I suggest we move to private messages so it will be more convenient for us to discuss the details)

Show original

Translate