Matviy is an excellent specialist and a very pleasant person to communicate with. The second stage of the project was completed with high quality and within the agreed deadlines. Throughout the work, he was always in touch, promptly answering questions, explaining technical details thoroughly, and offering solutions for complex tasks. I would like to particularly note his openness to feedback and willingness for constructive dialogue. All requests and edits were discussed without any issues, and during the work, there was a noticeable interest in achieving a quality result rather than just formally completing the task. We are satisfied with the collaboration. We continue to work on the next stages of the project and I can recommend Matviy as a responsible and competent developer.
32
0
Budget: 27000 UAH Deadline: 10 days
For the second stage, we can start within about 10 working days, but I would tie the 95% accuracy not to a feeling, but to a test sample with manual annotation and rules for controversial cases. Otherwise, it will not be an engineering acceptance, but guessing on transcripts =/
Look, there’s a nuance - first, we will check the results of the first stage, the quality of the texts, the structure of the data, access to the server, and how the records are already entering the system. After that, we will assemble the analysis module according to your logic, create a scoring from 0-100, classification of cold / medium / warm, separate features for needs, objections, deal stage, and manager's work. Next, calibration on your sample, so that the model does not provide a nice general summary, but meets the specified criteria.
We need 2 clarifications before starting:
> are there already 30-50 ideally annotated calls for accuracy checking?
> does the first stage currently only provide text or is there already a link between call - manager - lead - deal - department?
Similar cases from Ingello:
> https://business.ingello.com/vorfahr - AI and automation of business processes, close in logic to data processing and solutions
> https://business.ingello.com/forma-crm - CRM and management logic for process control
> https://business.ingello.com/platforma - corporate platform with roles, data, and scaling
Main page for systems - https://systems-fl.ingello.com
The budget of 27,000 UAH for the second stage is accepted as a working guideline, but !!the 95% guarantee can only be confirmed after reviewing the annotation and the quality of the transcription!!. Let's not complicate things - if the base of the first stage is normal, we will create a careful calibratable system, not a showcase for presentation.
6
0
Budget: 27000 UAH Deadline: 10 days
Good afternoon, Yuri.
The task is no longer just AI integration, but a full-fledged production system for call analysis with strict business logic for scoring and high load. I have experience in developing such pipelines: audio transcription, LLM analysis of dialogues according to specified rules, lead scoring, and integration with CRM under high load.
To accurately assess the current stage, I need to clarify:
What has already been implemented in stage 1 and what stack is being used?
Where are the calls currently stored and how is the flow from Ringostat organized?
How is the integration with amoCRM currently implemented?
What I propose: I will join stage 2, implement call analysis and scoring strictly according to your logic, calibrate the model to real data until achieving accuracy ≥95%, and ensure the recording of results in amoCRM. After that, I will transfer everything to GitHub with documentation and readiness for scaling.
I suggest we discuss the details!
5
0
Budget: 27000 UAH Deadline: 7 days
Hello, I worked on an AI call analysis system for a car dealer with 80+ managers - integration with CRM, transcription of 15,000+ calls/month, lead scoring with 97% accuracy. We increased conversion by 34%.
What specific scoring logic will be applied for the automotive business - are the features of imported cars and regional differences of clients taken into account?
I suggest we get in touch; I will provide you with a free technical consultation and we can outline a development plan + I will tell you about my team! ⚡
Matvii Lipnevsky
Winning proposal
8
0
Budget: 27000 UAH Deadline: 9 days
Good day. I will complete it on time and with quality. Message me privately, and we will discuss the details.
9
1
Budget: 27000 UAH Deadline: 3 days
Hello! Your project has sparked my great interest. I am ready to start working immediately and ensure high-quality execution.
7
1
Budget: 27000 UAH Deadline: 1 day
Hello!
I am a Full-Stack Software Engineer with over 7 years of experience in developing websites, SaaS solutions, complex web platforms, and MVPs for startups - from idea and architecture to production and support.
I work not only as a developer but also with a focus on business logic, scalability, and long-term support of solutions. My portfolio includes examples of implemented projects of varying complexity.
Technology stack:
PHP (Laravel, Symfony, Yii2),
Frontend: JavaScript (Vue.js, React.js), HTML5, CSS3,
Databases: MySQL, PostgreSQL.
Regarding timelines and costs - at this stage, it is difficult to accurately assess them without clarifying the requirements. To form a reasoned estimate, it is usually necessary to understand:
– project goals;
– key functionality;
– expected load and integrations;
– priorities (quick launch vs scalability).
I suggest starting with a brief clarification of requirements or a call / correspondence - after that, I will be able to provide a realistic estimate on timelines, budget, and implementation options.
I would be happy to discuss the details.
0
0
Budget: 27000 UAH Deadline: 10 days
There is already a practically ready solution for AI analytics of calls and lead scoring, which can be quickly adapted to your logic and launched in stage 2, we are in touch here in the correspondence =)
Stage 2 can be taken for 10 working days and 27,000 UAH, if the first stage consistently provides call texts, recordings related to leads, and there is access to examples for verification.
An important nuance regarding 95% - such accuracy can only be correctly calculated on a reference sample, where the correct answers are pre-labeled according to your criteria.
Therefore, I would not do a general AI analysis, but a verifiable evaluation model - rules, prompts, result scheme, re-validation, error log, and calibration on test calls.
For the implementation of stage 2, I see it this way:
- We take the call texts from the results of stage 1
- We bring them to a unified format
- We set up the analysis of needs, objections, deal stages, agreements, and the quality of the manager's work
- We calculate the call score from 0 to 100 and classify the lead
- We prepare the data structure so that stages 3-5 can be smoothly integrated into amoCRM, signals, and dashboard
- We transfer the code and materials to the company's private GitHub repository
To start, we need access to the current server, repository, or files from stage 1, an example of 20-50 calls, scoring criteria, and preferably 10-20 already labeled reference examples.
I will clarify 2 points to avoid guessing with the server, I can sometimes complicate life for no reason =/
- What format does the first stage currently provide transcripts in - database, files, API, or another storage
- Is there already manual labeling of calls for 95% verification, or does it need to be done together as part of the calibration
Similar works by Ingello:
- https://business.ingello.com/vorfahr - AI automation and agent logic for workflows
- https://business.ingello.com/forma-crm - CRM and management logic for operational teams
Main page for projects on the marketplace:
https://systems-fl.ingello.com
