STULINK

We’re a fast-growing startup working on a next-generation conversational platform that blends natural voice, real-time interaction, and emotional intelligence. The product is already live in private beta — we’re looking for an AI Engineer who can refine, optimize, and scale our current model integrations. Responsibilities Build, fine-tune, and deploy LLM-based conversational agents using OpenAI, Gemini, or similar APIs Optimize real-time audio pipelines , including transcription (STT) and speech synthesis (TTS) Develop and test prompt-engineering logic and dynamic response generation Collaborate with backend engineers to improve latency, accuracy, and contextual memory Implement custom tools and APIs for real-time reasoning and emotion detection Monitor and analyze model performance, create feedback loops for continuous improvement Requirements Proven experience integrating OpenAI, Anthropic, Gemini, or similar LLMs Strong coding skills in Python / Node.js Understanding of NLP, embeddings, prompt design, and text generation pipelines Hands-on experience with real-time systems (WebSocket, LiveKit, Twilio, etc.) Ability to debug and optimize model response times and context flow Solid foundation in API integration, cloud deployment, and model testing Nice to Have Experience with emotion recognition , voice-to-voice interfaces , or persona-based AI Familiarity with LangChain , RAG , or vector databases Experience in fine-tuning smaller open-source models (LLaMA, Mistral, etc.) Previous work on AI companions , counseling bots , or real-time agents What We Offer Competitive pay (monthly or milestone-based) Opportunity to work on cutting-edge real-time AI products Fast, creative environment — direct impact on user experience Long-term potential for growth with a core founding team ⚠️ Note: This project is currently in private beta. Full product details will be shared under NDA with shortlisted candidates.