We are looking for a Technical AI Assistant to develop, test, and debug internal company systems together with the owner.
This is not a classic programmer vacancy. We don't need someone who writes all the code manually and knows programming languages by heart. What's more important is technical thinking, attention to detail, patience, and the ability to use AI tools to understand how a system works and why it sometimes produces incorrect results.
What has already been created in the company:
- Proprietary CRM system;
- AI evaluator for industrial equipment cost;
- Digital controller for deals and manager tasks;
- Call transcription and analysis;
- Manager dashboards;
- Automation of individual sales and procurement processes.
All these systems are already in use in real work. They are constantly evolving, so we need a person who can gradually take over their testing, debugging, and technical support.
What you will need to do:
- Receive task descriptions and expected results from the owner;
- Understand existing processes and system logic;
- Find the causes of incorrect results and technical errors;
- Use AI tools to make changes to code, settings, model queries, and integrations;
- Verify changes with real historical examples;
- Create tests and ensure that the error does not repeat;
- Ensure that fixing one process does not break others;
- Briefly document what was changed and how it was verified;
- Gradually take over support for individual parts of the system.
Important feature of the work
A significant portion of the tasks involves not quickly creating something new, but bringing an already created system to a stable result.
For example, you will need to figure out why the AI evaluator incorrectly determines the price in three cases out of a hundred, find a common cause, fix it, and check that other results have not worsened after the change.
If you enjoy finding the causes of errors, testing hypotheses, and bringing results to accuracy — this job might be for you. If you are only interested in quickly creating prototypes and moving on to the next task — it will likely be difficult.
What we expect:
- Technical and systematic mindset;
- Ability to systematically understand an unfamiliar system;
- Attention to small errors and inconsistencies;
- Ability not to trust the AI's first answer, but to verify the result;
- Experience with ChatGPT, Claude, Codex, Cursor, or similar tools;
- Readiness to work with n8n, API, logs, JSON, and code with the help of AI;
- Independence after receiving a task and necessary context;
- Ability to explain in simple terms what was done and how it was verified.
Programming experience will be a plus, but is not mandatory. We may also consider a junior tester, integrator, system analyst, or automation specialist.
Conditions:
- Permanent remote work;
- Full-time employment;
- Income 60,000–80,000 rubles depending on skills and independence;
- Direct work with the company owner;
- Gradual immersion into the system, starting with one limited process;
- Paid practical assignment before employment;
- Opportunity to grow into the owner's primary technical assistant and take over support for a significant part of the system.
To apply, please answer four questions:
- What have you already created or configured using AI tools?
- Provide an example of an error whose cause you had to find yourself.
- What would you check if the system correctly calculates 97 out of 100 prices, and overestimates the remaining three by ten times?
- What income level are you considering?
You can attach links, screenshots, or a brief description of your own projects. We value real examples and a way of thinking more than a formal resume.
