Low-Code Prototype
Implement prototypes of AI agent functions on the constructor.
Since the project is at the MVP stage, it is important for us to quickly test ideas, so we are looking for a specialist who can build prototypes on low-code platforms (Make, n8n, Zapier or similar).
Tasks 📋
Prototype assembly
Create working scenarios on low-code platforms (Make/n8n/Zapier) without "manual" coding.
Write prompts (system, user, few-shot) and connect external LLM-APIs (OpenAI, Anthropic, etc.).
Token and latency optimization
Configure trimming/summarization of history so that the bot responds quickly and does not "consume" unnecessary tokens.
Monitor response quality: relevance, tone, absence of hallucinations.
Memory management
Configure short-term and long-term memory using platform tools (for example, Storage modules in Make/n8n).
Decide what to save and what to delete to avoid "drowning" in the context.
What you will need 💡
Experience with Make, n8n, Zapier or similar
Ability to write clear, effective prompts for GPT-4/Claude/Llama.
Understanding of basic LLM concepts: tokens, temperature, system-prompt, context length.
AI assistant tasks: 🚀
Ask questions to gather reports from employees
Analyze 2 opinions for compliance
Analyze reports