In this article, we showcase an AI Interviewer developed at ELEKS to accelerate technical interviews for software engineers. By automating initial screening rounds, companies can reduce time-to-hire and allow hiring managers to focus on the most qualified candidates.
Candidates can also use this technology to practice their interview skills. The AI agents were designed to listen and respond in a human-like manner. Below, we share insights from our team on how we made this possible.
- Our team built an AI interview platform to automate technical screening for software engineers, reducing time-to-hire.
- Two voice-based AI agents conduct realistic interviews, adapting dynamically to candidate responses.
- The system generates detailed feedback reports scoring technical knowledge, delivery, and communication gaps.
The challenge: creating an AI interview experience that feels authentic
AI Interviewer is an intelligent platform that allows users to converse with realistic virtual interviewers — AI agents trained to recreate actual interview scenarios and analyse both responses and presentation style of the candidate.
The outcome is a digital experience that feels natural and provides clear, actionable feedback to help candidates progress with every session.
The solution: AI agents that interview, adapt, and provide feedback
To start a session, the user fills out a form with the interview duration, project description, and uploads a CV. The interview is conducted by two AI agents. The first, a welcoming coordinator, introduces the interview process and guides users through initial general and behavioural questions. After these, the session concludes as the first agent summarises and directs users to their feedback report.
The second agent is a technical lead. He adapts dynamically to user input, providing clarifying questions and technical challenges to simulate a rigorous engineering interview.
These agents are powered by GPT-based models with unique prompts and behaviour settings to balance empathy and technical depth. Azure OpenAI GPT models are deployed on Azure AI Foundry to handle the conversation logic.
Users speak through their microphone and receive immediate responses from the AI Interviewer. Using LiveKit and Azure Speech Services, the platform transcribes spoken input in real time and converts the AI's replies into human-like audio.
The platform generates a feedback report that assesses both content and delivery. It evaluates metrics such as syntax accuracy, depth of understanding, completeness of responses, and identifies any missed topics.
Following the interview, the system generates a summary, the full conversation transcript, and detailed scores for each category. The report includes specific, actionable advice — such as recommending measurable metrics to fix hesitation during project explanations.
AI Interviewer interface: interview setup form
Live interview with animated avatars
Post-interview feedback report
What powers the AI interviewer: Azure-based architecture and technology stack
Azure’s built-in artificial intelligence and data tools help create secure, high-performing solutions. With strong compliance and security, identity management, and data privacy controls, the platform delivers reliable performance in every deployment. This makes it a great choice for universities, recruitment platforms, and corporate training.
This system uses LiveKit for real-time audio and video. Additionally, the Silero Voice Activity Detection (VAD) model allows the AI to identify the exact moment when to listen and when to respond, so users can speak without interruptions.
GPT-based models are accessed via Azure OpenAI Service to support seamless language processing and context-aware dialogue. These models deliver tailored responses and generate objective post-interview summaries. Together, these components create an adaptive environment that combines real-time interaction with a robust cloud infrastructure for a professional interview experience.
Testing phase and further development
The AI Interviewer is currently in the testing phase phase. We continue to collaborate with corporate partners to test the conversation quality and feedback mechanisms. Based on this input, we redefine AI personas and dashboards to prepare them for a broader launch.
The technology streamlines hiring by conducting consistent, unbiased screenings at scale. Companies can evaluate more candidates quickly, reduce interviewer workload, and maintain assessment quality.
The roadmap moves toward a sophisticated model driven by artificial intelligence that uses multiple input streams. Facial expression analysis and communication pattern tracking will further boost assessment accuracy, helping clients find top talent faster.
FAQs
An AI chatbot responds to your questions. You ask, it answers. It's reactive, waiting for your input before replying. Most customer service bots on websites are chatbots.
An AI agent pursues goals and takes initiative. It doesn't just answer; it plans, adapts, and acts to accomplish specific tasks. An agent can break down complex problems, make decisions, use tools, and adjust its approach based on results.
AI conversational assistant is software powered by artificial intelligence that understands and responds to your questions and requests in natural language. AI assistants help with answering questions, writing or editing documents, solving problems, performing calculations, searching for information, and much more. Popular examples include Claude, ChatGPT, Siri, Alexa, and Google Assistant.
They use advanced machine learning models trained on large datasets to understand context and generate useful responses across chat, voice, mobile apps, and web platforms.
Yes, to some extent. AI can copy how people talk, their tone of voice, and conversational style. Modern AI can sound natural, remember what you said earlier, and adjust its personality based on how it's programmed.
AI can mimic the way someone speaks or writes, their tone and personality, expertise in specific topics, and natural conversation flow. However, it cannot replicate real emotions or consciousness, actual life experiences, true understanding, or human intuition and moral judgment.
Related Insights
The breadth of knowledge and understanding that ELEKS has within its walls allows us to leverage that expertise to make superior deliverables for our customers. When you work with ELEKS, you are working with the top 1% of the aptitude and engineering excellence of the whole country.
Right from the start, we really liked ELEKS’ commitment and engagement. They came to us with their best people to try to understand our context, our business idea, and developed the first prototype with us. They were very professional and very customer oriented. I think, without ELEKS it probably would not have been possible to have such a successful product in such a short period of time.
ELEKS has been involved in the development of a number of our consumer-facing websites and mobile applications that allow our customers to easily track their shipments, get the information they need as well as stay in touch with us. We’ve appreciated the level of ELEKS’ expertise, responsiveness and attention to details.