very good communication, Arthur is polite and professionally performs his job. I will definitely reach out to him again.
1 287
1
Budget: 1000 UAH Deadline: 1 day
Hello. I am ready to conduct an audit. Can I review the code to assess the task in terms of deadlines and price?
2
1
Budget: 15000 UAH Deadline: 42 days
Hello!
I carefully read the specifications — I understand what is behind each block. This is the construction of a stable AI product that can be trusted: songs without fabricated facts, orders that do not get lost, a system that withstands load.
The main problem is product-related: GPT fabricates facts, songs are template-based, orders get lost during failures. We need a product that takes ONLY real client data, verifies each generation, and works stably under load.
HOW I WILL DO IT
Stage 1 — Foundation: refactoring architecture, separation into layers (Telegram / Payment / GPT / Music / Queue / Admin / Debug), Redis queues, payment protection, database optimization.
Stage 2 — Quality: new GPT without template phrases, Fact Check module with automatic regeneration for fabricated facts, Music Brief AI between text and Mureka, quality assessment before sending to the client.
Stage 3 — Music: optimization of Mureka, integration of Suno as a second generator, A/B testing of both services.
Stage 4 — Infrastructure: Debug system with a complete chain for each order, admin panel with regeneration, stress tests for 1000 simultaneous orders.
STACK: Python 3.11, Aiogram 3, PostgreSQL, Redis, GPT-4o, Mureka + Suno, Docker.
QUESTIONS BEFORE STARTING
1. Is there already a bot code or are we starting from scratch?
2. What API keys are available? (OpenAI, Mureka, Suno, payment system)
3. What payment system is being used? (LiqPay, Fondy, other)
4. Where will the server be deployed?
BUDGET BY STAGES
Stage 1 — $300–400 / 1–1.5 weeks
Stage 2 — $400–600 / 1.5–2 weeks
Stage 3 — $300–400 / 1–1.5 weeks
Stage 4 — $300–400 / 1–1.5 weeks
Total — $1300–1800 / 5–6 weeks
We can start with Stage 1 — you see the result, then we continue. Write to me — let's get into specifics.
Artur Boiko
Winning proposal
5
0
Budget: 18500 UAH Deadline: 10 days
Good day!
Engineering AI pipelines for generation is our core focus, so the specifications read like "our own." Here, the essence is not in "bot corrections," but in the quality construction of the pipeline: anti-hallucination prompting, fact-check layer, intermediate Music Brief, eval loop for quality control, and fault-tolerant queues. These are the systems we work with.
As we see, the key blocks are:
GPT V2 + Fact Check (Blocks 2-3): reworking system prompts, cleaning up template AI phrases ("neon lights," "alleys pulsating," etc.) and a separate fact-check layer that catches fabricated cities/events/facts and sends them for regeneration. If there is little data from the client, we generate a simple universal song without fabrications. This is the right approach: the model should not invent what is not in the brief.
Music Brief AI (Block 4): the intermediate layer GPT → Music Brief → Mureka really enhances quality — the generator receives a structured brief (genre, tempo, dynamics, vocals, climax), not raw text.
Premium Quality (Block 14): this is the eval loop — AI evaluates the text, weak results are regenerated to meet the quality threshold, the client receives the best version, not the first random one.
Architecture + Reliability (Blocks 1, 5, 8-13): division into layers (Telegram / Payment / GPT / Music / Queue / Admin / Debug), idempotency of payments (without loss of orders and duplicate charges), recovery of stalled orders, Redis queue control, debug log of each generation, stress test at 100/500/1000.
Suno (Block 6): multi-generator under a common interface Music Engine + A/B Mureka vs Suno.
Approach: the project is large, so we will start with an audit of the current code, then a plan by blocks and work in stages. Critical first (loss of orders after payment, stalled queues, fact-check), then quality and Suno. We do not rewrite everything at once.
We are ready to start the audit and propose a roadmap. Details — in private messages.
14
0
Budget: 27000 UAH Deadline: 15 days
Hello, I have experience in developing similar AI products, refactoring architecture into modules, integrating with GPT/music generators (Mureka, Suno), building queues on Redis, payment systems, admin panels, and debugging systems, so your task is clear to me and I am ready to complete it. The deadline is 15 days, the cost is 30,000 UAH. I hope for cooperation!
4
0
Budget: 1000 UAH Deadline: 1 day
Hello!
Our team is ready to implement the "turnkey" project — from design to full development and launch. We have experienced developers, designers, and UX/UI specialists in our team, so we can create a convenient, modern, and functional product according to your requirements.
We will be able to determine the timelines and budget accurately after discussing the details and understanding the full scope of work. I would appreciate your feedback, and I can send examples of our work in private messages.
4
0
Budget: 12000 UAH Deadline: 14 days
It is important not just to "complete the bot," but to eliminate the bottlenecks through which personal song generation breaks down at the intersection of Telegram, payments, GPT, queues, and the music pipeline. I would start with a brief audit of the current logic and specifications to immediately identify critical scenarios: where orders get lost, where errors occur after payment, and how to control the quality of the output without manual chaos. As a result, you will receive not just a set of specific corrections, but a clear plan for further development with priorities that can realistically be implemented without unnecessary rework and the risk of breaking functional parts. If convenient, I can immediately help structure the scope of the first phase of work and from this accurately assess timelines and budget.
Budget: 10000 UAH Deadline: 5 days
Hello! Our technical team has been developing complex bots and integrating intelligent solutions into business processes for over 4 years. We understand perfectly how to properly structure technical specifications for AI projects so that developers do not make mistakes during implementation. We will translate your ideas and business requirements for personal generation into clear technical task language. We will outline user behavior scenarios, parsing algorithms, and input data processing for AI, as well as the logic for forming the final result. We will definitely establish criteria for evaluating the quality of responses from the neural network to minimize bot errors. You can familiarize yourself with our interface design and system logic standards through examples of our work: farfieworldwide.com, freshagro.com.ua, espressolab.com.ua. Send us a private message describing your wishes, and we will immediately prepare a plan for structuring the future technical specifications.
20
0
Budget: 8000 UAH Deadline: 14 days
Hello. I understand the task: we need to turn a working but confused bot into a stable, scalable product. This means separating the logic into distinct layers, such as Telegram, payment, GPT, music generation, queue, admin panel, debugging, rewriting the text generation so that GPT does not invent facts about a person, adding a separate Fact Check between the questionnaire and the brief, inserting the Music Brief AI stage before Mureka, connecting Suno as a second generator, and cleaning up the queues, payments, and debugging for each generation.
Regarding architecture, I would create a clear division into services with narrow interfaces between them, so that the queue, payment, and generation do not depend on each other and each layer can be tested separately. For GPT v2, I would rewrite the system prompts under a strict contract: the model only works with what is in the questionnaire, and if there is little data, it returns a simple universal song without invented cities, events, and professions. Template phrases like neon lights would be removed both through instructions in the prompt and through a post-filter based on a stop list, so they do not pass even by accident.
I would make Fact Check a separate step that verifies the generated text against the questionnaire and returns a decision on whether it passed or needs regeneration, with automatic retries for any found fabrications. Music Brief AI would form a structured brief based on genre, tempo, emotion, dynamics, vocals, instruments, climax, and then send it to Mureka or Suno. For queues and orders, I would add control for stuck tasks, recovery after failures, and payment protection to ensure that orders are not lost after payment.
I work a lot with LLM in production: prompt engineering, hallucination control, and tool use on a voice AI assistant, plus bots in Python with Redis queues and a payment layer. The question is: is the bot currently on aiogram or something else, and can I get access to the code for auditing to accurately assess the scope by blocks? I am ready to start with a map of the current architecture and a plan for breaking it into modules.
21
0
Budget: 20000 UAH Deadline: 30 days
Hello. I can complete your task. I have experience. Write to me and we will agree.
1
1
Budget: 9000 UAH Deadline: 1 day
Good day. I am ready to complete this project as I have extensive experience in development.
14
0
Budget: 15000 UAH Deadline: 21 days
Good day, Olga!
The technical specification reads as our own - we understand each block from the inside: anti-hallucination prompting, Fact Check layer, Music Brief between GPT and Mureka, eval loop for quality control, fault-tolerant Redis queues, payment protection.
The main problem is understood and familiar: GPT invents facts, songs are template-based, orders get lost during failures. The client should receive a song based on real data, not the first random generation.
I suggest a phased approach through Safe - it's comfortable for both you and us: each phase is reserved separately, and you see the result before paying for the next.
After the code audit, we will clarify the timeline and cost.
When is it convenient to discuss the start?
0
0
Budget: 1000 UAH Deadline: 3 days
Hello!
We can help with the refinement of the AI bot for generating personalized songs.
The task is clear: we need to not just "improve the prompts," but stabilize the entire system — architecture, GPT logic, fact check, music brief, generators, queues, payments, debugging, and quality control.
We can take on the following:
— refactoring the architecture by modules
— updating the GPT generation pipeline
— Fact Check AI for verifying fabricated facts
— Music Brief AI before passing it to Mureka / Suno
— optimizing the Mureka API
— integrating Suno as a second music engine
— debug system for each order
— monitoring statuses, stalled orders, and restarts
— Redis / queues / background tasks
— checking payment logic
— admin panel for controlling generations
— stress test and premium quality control system
The project is better implemented in stages: first, an audit and stabilization of the current system, then GPT/Fact Check/Music Brief, after that Suno, debugging, quality control, and stress testing.
We are ready to familiarize ourselves with the current architecture, code, Mureka/Suno API, and after the audit, propose a work plan, timelines, and budget.
11
0
Budget: 11111 UAH Deadline: 15 days
I will start refactoring the architecture of the AI bot, separating the logic into individual modules. I am including the Telegram Layer, Payment Layer, GPT Layer, Music Generation Layer, Queue Layer, Admin Layer, and Debug Layer in preparation for scaling. I will implement a fact-checking module that regenerates songs upon detecting fabrications.
Portfolio - 
Freelancehunt
41
4
Budget: 1000 UAH Deadline: 3 days
Good day! I professionally develop bots in Telegram and web applications in Python. I am ready to effectively enhance your AI bot for generating songs.
Please message me privately to discuss the details. I have significant experience with APIs (GPT, Suno, Mureka), architecture, Redis, databases, and LLM.
10
0
Budget: 3000 UAH Deadline: 3 days
Good day!
I have significant experience in developing and scaling AI products, building complex backend systems, integrating LLM models, Telegram bots, payment systems, processing queues, as well as high-load services and multi-step content generation pipelines.
In your specifications, I see not just a set of individual tasks, but a full evolution of the product from MVP to a stable production-ready system. I particularly value the emphasis on the quality of generations, fact-checking, debugging, and scaling — these blocks often determine the success of AI services in the long term.
To implement this, I am ready to conduct a comprehensive audit of the current architecture and perform refactoring by isolating separate layers: Telegram, Payments, GPT, Music Generation, Queue Management, Admin, and Debug System. This will significantly simplify the support and further development of the project.
I also have experience with:
• OpenAI and other LLM models;
• building multi-step AI pipelines;
• Redis, queues, and background tasks;
• webhook and polling architectures;
• integrations with third-party APIs;
• logging and monitoring systems;
• PostgreSQL and database optimization;
• load testing;
• Telegram Bot API;
• payment integrations.
I would pay special attention to the implementation of the Fact Check Layer, AI Quality Control, and Debug System, as they allow for controlling the quality of generations, eliminating model fabrications, and ensuring complete transparency of each order from the questionnaire to the final result.
I am also ready to implement a universal Music Engine with support for multiple generators (Mureka, Suno, and future services), which will allow for easy A/B testing and avoid dependency on a single provider.
I am prepared to thoroughly familiarize myself with the current codebase, infrastructure, and propose an optimal implementation plan. We can discuss the details in private messages.
Budget: 2000 UAH Deadline: 1 day
Hello
interesting task
I have experience in this field
I am ready to discuss everything in private
3
1
Budget: 7500 UAH Deadline: 10 days
Hello! I did an audit for you, ready to take on the job!!!!!!!!!!!!!!!!!
Budget: 10000 UAH Deadline: 1 day
Hello!
We can provide you with a solution for the enhancement of the AI bot and its generation.
1. Which blocks are more important to close first?
2. Do you have the current code and access to it?
—
About us
We are dZENcode – a full-cycle IT development company: from design and programming to integrations and post-release support of digital solutions. We create projects from scratch, as well as connect to existing solutions that need enhancement, development, or technical support.
You can find detailed information about our services and rates on our official website: Freelancehunt
Take a look – after that we can discuss the details and agree on the next step.
⚠️ After clarifying all the details, we will determine the scope of work, the suitable format of cooperation: task-based, outsourcing, or outstaffing – and the final cost.
With us, projects are guaranteed to reach release:
• 10+ years providing IT services;
• 90+ in-house specialists;
• 250+ public reviews since 2015;
• We support products under SLA after launch;
• We work under NDA and a contract with the company!
13
0
Budget: 25000 UAH Deadline: 7 days
Hello, I work a lot with AI bots and can implement yours without any problems. To avoid writing a long message about the stages, I will write to you in private messages.
git: https://github.com/onyx144
Example: https://217-154-170-186.nip.io/
9
1
Budget: 2000 UAH Deadline: 3 days
Hello! Your project has sparked great interest in me. I am ready to start working immediately and ensure high-quality execution.
The list does not show proposals concealed by the client or freelancer with a Plus profile, as well as proposals violating rules
