Priyank Jhaveri

About

Currently at Uplift AI shipping production LLM features that generate personalized growth insights from user reflections using BERT + embeddings + RAG, with strong safety/guardrail practices for sensitive contexts. Also built an end-to-end React Native UGC challenge submission/moderation system that improved repeat submissions and 7-day retention, and has applied rigorous clinical-style evaluation methods on a dental X-ray disease detection project to reduce false negatives.

Key Strengths

• End-to-end feature ownership across mobile app + API + database
• Data-driven product iteration using funnel instrumentation and support feedback
• Risk-aware design for UGC (manual moderation to mitigate spam/legal/trust issues)
• Built production RAG workflow with embeddings and grounding requirements
• Implemented safety guardrails for sensitive/mental-health-adjacent LLM outputs (profanity/self-harm detection, escalation messaging, human review)
• Designed robust ML evaluation loops with stratified real + synthetic test sets and edge-case focus
• Reduced clinically risky false negatives via targeted augmentation and regression testing on frozen holdout set
• Built scalable ingestion pipelines with reliability patterns (idempotency, retries, DLQ) and observability

Reference Highlights

• Owned and delivered multiple MVP-critical features for planned launch
• Co-team lead with leadership responsibilities
• Effective at managing team resources and tasks
• Collaborative problem-solving and escalation to resolve blockers quickly
• Solid full-stack understanding of system decisions and architecture
• Works well cross-functionally with architect and Dev/Sec/Ops

Experience

Education

Skills

Languages

Publications

1 publication