AI Intern

SViam is building an autonomous AI interviewer that conducts real time voice and coding interviews, remembers context across rounds, learns each company's hiring bar, and catches cheating with the most advanced proctoring system in the industry. We are replacing the broken hiring pipeline, not patching it.

You will work directly with the founder. No layers, no meetings, no busywork. Just building.

1. Work on SViam's AI interview engine, the core product that conducts live adaptive interviews.
2. Build and improve proctoring signals like typing cadence analysis, behavioral scoring, and integrity intervention logic.
3. Experiment with LLM prompting strategies for adaptive follow up questions, scorecard generation, and candidate evaluation.
4. Contribute to the calibration engine that learns a company's hiring bar from their past decisions.
5. Write Python services that process real time WebSocket data during live interviews.
6. Help design the question bank and difficulty progression system across interview rounds.

1. Strong fundamentals in Python. You can write clean, working code without hand holding.
2. Some exposure to LLMs, whether that is prompting, fine tuning, RAG, or building with APIs like OpenAI or Claude.
3. Curious about how AI can replace broken hiring processes. You have opinions about what is wrong with interviewing today.
4. You are a 3rd or 4th year college student (or recently graduated) looking for real work, not a certificate.
5. You ship fast. You do not wait to be told what to do next.

1. Direct access to the founder. Daily async communication, weekly syncs.
2. You will push to production in week one. This is not a training program.
3. Flexible hours, fully remote. We care about output, not when you are online.
4. Real ownership. Features you build will be used by companies running live interviews.
5. A reference and recommendation letter at the end if the work is strong.

The base internship is 2 months. A 6-months internship is also available for candidates who want deeper involvement. Extension decisions are made at the end of month two based on performance and mutual fit.