AI Engineer Offline

Impressit is looking for an AI Engineer to join a cross-functional product team working on an AI-native EdTech platform focused on workplace skills assessment, personalised coaching, and voice-based practice experiences for young adults.

What do we expect?

— 2–3+ years of commercial experience building production AI/ML systems using LLMs or SLMs
— Strong Python skills with hands-on experience in FastAPI and Pydantic
— Data-driven mindset: experience designing experiments, collecting evaluation data, and iterating based on measured outcomes
— Hands-on experience with Google Cloud Platform, especially Vertex AI
— Experience integrating speech-to-text services (e.g. Google Speech-to-Text, Assembly AI)
— Solid experience with audio processing and vocal analysis
— Experience building AI pipelines for batch inference
— Familiarity with prompt management and AI observability tools (e.g. Langfuse)
— Understanding of evaluation approaches for AI/ML systems
— Experience with SQL databases, particularly PostgreSQL
— Comfortable working with Git and Docker
— Strong problem-solving skills and ability to work autonomously
— Good communicator, comfortable collaborating in cross-functional teams
— Upper-Intermediate or higher level of English

What will you do?

— Build and maintain production AI pipelines for assessment, feedback, and coaching
— Design and iterate on prompt and context engineering for scenario generation and performance evaluation
— Evaluate and select LLMs and SLMs for different AI tasks (transcription, analysis, generation, assessment)
— Develop AI agents and role-playing systems for interactive learning experiences
— Build pipelines for audio processing and transcription analysis
— Implement batch inference workflows for scalable assessment scoring
— Design evaluation frameworks using human-labelled data and quantitative metrics
— Collaborate with product and engineering teams on AI-driven features
— Contribute to AI architecture decisions and long-term technical strategy

As a plus:

— Experience with multi-agent orchestration frameworks (e.g. LangGraph, Google ADK)
— Familiarity with Python-based prototyping tools (e.g. Streamlit)
— Previous experience in EdTech, digital coaching, or learning platforms
— Hands-on experience with audio processing libraries (e.g. librosa)
— Passion for evidence-based AI, system reliability, and end-user quality