AI Personality System Architect
Architect long-horizon persona consistency, activation steering, emotional tone stabilization, and ethical value alignment across autonomous agent fleets.
Role Overview & Operational Scope
As AI Personality System Architect at Cehpoint, you will define the psychological fidelity and long-term behavioral integrity of our autonomous agent fleets. You will prevent persona drift, ensure high-stakes commercial alignment, and design mathematical steerability frameworks that govern how agents interact with enterprise executives.
Key Responsibilities & Production Deliverables
- Engineer dynamic persona vectors and latent activation steering mechanisms across open and frontier transformer models.
- Develop automated evaluation suites to benchmark persona stability, tone drift, and emotional modulation across 100+ turn sessions.
- Design epistemic boundaries preventing agent hallucination of personal knowledge while maintaining rich persona fidelity.
- Integrate personality layers with episodic memory stores to deliver seamless, historically grounded agent interactions.
- Collaborate with red-teaming teams to patch persona bypass exploits and jailbreak attacks targeting identity layers.
Mandatory Foundational Knowledge
- Deep understanding of transformer internal representations, residual stream steering, and mechanistic interpretability.
- Familiarity with RLHF, DPO, and representation engineering methodologies (e.g. contrastive activation addition).
- Strong knowledge of psychometrics, conversational turn dynamics, and state-machine persona constraints.
Mandatory Practical Skills & Architecture
- Advanced Python engineering with PyTorch, Hugging Face Transformers, and vLLM inference engines.
- Experience building low-latency token streaming pipelines with deterministic guardrail layers.
- Demonstrated track record of designing robust system prompts, steering vectors, and evaluation harnesses.
Problem Solving, Execution Rigor & Curiosity
- Obsession with the frontier of mechanistic interpretability and cognitive science.
- Refusal to settle for simplistic prompt engineering when structural representation steering is required.
- Exceptional discipline in benchmarking and empirical reproducibility.
5-Day Live Technical Evaluation Milestone
5-Day Live Practical Milestone: Design and deploy a deterministic persona-anchored agent pipeline that maintains strict stylistic and ethical boundaries across 50 adversarial multi-turn conversation probes without persona collapse (strictly 5 working days). Verification unlocks immediate official corporate offer letter.
Institutional Hiring Protocol: Candidates who pass initial resume screening are invited to a live, practical evaluation milestone spanning strictly not more than 5 working days. Verifiable completion and code audit by your assigned senior engineering mentor is the sole prerequisite for official corporate offer letter issuance.
Compensation, Total Rewards & Advancement
- High-tier executive compensation (₹14,00,000–₹24,00,000) with equity participation options.
- Unlimited access to GPU clusters and frontier model compute credits.
- Fully remote environment with home laboratory hardware allowance.
- Direct publication sponsorship for novel alignment and steering discoveries.
Dedicated Inquiries Inbox for This Role
Have questions regarding architecture scope or wish to share private research repos directly? Messages sent to this address route straight to the engineering leads reviewing this opening.
ai-personality-system-architect-careers@cehpoint.co.in
Open Mail Client →