Fellowship · Remote (India)

AI Personality & Persona Research Intern

Cehpoint is hiring for this role: fellowship, Remote (India), ₹15,000 – ₹25,000/month + Full-Time PPO (₹12–18 LPA). Research activation steering vectors, long-turn persona consistency, and multi-turn behavioral benchmarking under senior engineering mentorship.

About the role

This is a high-intensity, project-based internship designed to turn promising students and graduates into elite AI research engineers. You will not fetch coffee; you will run experiments, author benchmarks, and contribute directly to production agent steering systems.

Join Cehpoint as an AI Personality Research Intern. You will work directly with our senior research team to evaluate persona drift, build prompt steering datasets, and test behavioral stability across autonomous agent interactions.

What you will do

  • Design and run multi-turn conversation probes to measure persona fidelity across different LLM backends.
  • Build synthetic evaluation datasets testing agent resilience against identity-inversion attacks.
  • Assist senior architects in tuning activation steering vectors and system prompt scaffolds.
  • Document experimental findings and maintain rigorous reproducibility logs in GitHub.
  • Present weekly research findings directly to engineering leadership.

What you should know

  • Foundational knowledge of Python programming, data structures, and basic object-oriented principles.
  • Conceptual understanding of modern LLMs, tokenization, embeddings, and attention mechanics.
  • Familiarity with Git version control and command line navigation.

What you should be able to do

  • Ability to write clean, readable Python scripts using standard data science libraries (Pandas, NumPy).
  • Hands-on experience making API calls to LLM endpoints (OpenAI, Anthropic, Gemini, or open-source).
  • Strong analytical writing skills for documenting bug reports and test results.

How you work

  • Genuine, burning curiosity about the inner workings of artificial neural networks.
  • Eagerness to receive rigorous code reviews and rapidly incorporate technical feedback.
  • High intrinsic motivation to build real-world software.

Skills we check

  • Python
  • PyTorch Basics
  • Prompt Engineering
  • Evaluation Harnesses
  • Transformers
  • Synthetic Data Creation
  • Git

What you get

  • Guaranteed paid monthly fellowship (₹15,000–₹25,000/month) with zero unpaid grunt work.
  • Clear PPO conversion track to full-time AI Engineer (₹12,00,000–₹18,00,000 CTC).
  • Direct 1-on-1 mentorship with senior AI architects and publication co-authorship opportunities.
  • 100% remote flexibility with verified corporate completion certificate.

How we select for this role

We read every application ourselves. Here is what happens after you apply.

  1. Apply with a short form

    Paste a link to your CV and answer five questions in your own words. No sign-in and no file upload. About fifteen minutes.

  2. Integrity check

    When you press Check my answers, an automated check confirms the answers were typed by you. You see the result straight away, before you submit.

  3. Choose your project date

    If no cheating is detected you are shortlisted, and we email you a link to choose the day your project starts.

  4. 5-day practical project

    On that day our team contacts you with the brief: real work, 5 working days, reviewed by an engineer. For this role: Build a testing harness that probes a designated agent persona across 20 distinct dialogue scenarios, scoring stylistic consistency and flagging deviations (strictly 5 working days). Top performers receive immediate PPO confirmation.

  5. Offer

    If your work solves the problem well, you receive your offer letter by email and in the portal.