Pro
Al Kaizar Sappayani
AI Engineer(AI Agents, LLM Apps, Chatbots, RAG)
New to Contra
Al Kaizar is ready for their next project!
Marketing Assistant Bot
As part of my work in AI-driven automation, I built a Marketing Assistant Bot using Relevance AI to address a major challenge in customer engagement: inefficient lead handling and inconsistent follow-ups. Businesses often struggle with manually qualifying leads, responding to competitor comparisons, and guiding potential customers through the sales funnel. The lack of automation resulted in lost opportunities, delayed responses, and high drop-off rates. To solve this, I developed an assistant on Relevance AI
19
122
AI Sales Agent
Problem: Customers needed fast, accurate answers on courses/licensing/policies from a huge, changing catalog; support handled repetitive questions and compliance is state-specific.
Solution: Built RUBI, an embedded RAG purchase advisor in using ETL + multimodal indexing and hybrid retrieval (metadata + vector + structured lookups) with cited, grounded answers, purchase flows, analytics, and tuning loops.
Metrics: 695,042 conversations; 2.6 Qs/convo; feedback 6% up / 94% down; ~10s to first token, ~11s complete; KB >100k structured rows.
2
13
AI Generated Data Analysis
Problem: Weekly call-center ops data needed actionable synthesis; manual analysis was slow and inconsistent.
Solution: Built an LLM data analysis agent: ingest call data, run NLP + aggregations, detect anomalies, and generate a weekly executive PDF with insights, metrics, and recommended actions Metrics Automated weekly PDF; similar setups cut analysis time ~90% and improved decision turnaround (one related client saw 17% revenue lift over 7 months)
Tech: Python ETL, Pandas, NLP, GPT-4o, Airflow
Result: Consistent reports that surface issues fast and drive quicker corrective action.
13
100
AI - powered insurance Agent
Problem: Prior auth was manual—staff reviewed patient docs to predict approvals, making the process slow and resource-heavy.
Solution: Built a HIPAA-compliant AI auth system: extract clinical data (OCR + clinical NLP), form a structured EHR-like record, evaluate insurer rules, and generate approval suggestions; added voice input workflows via STT/TTS plus secure storage and audit logging.
Metrics: Reduced turnaround time and improved SLA adherence (similar projects reached ~98%). Result: Automated major parts of auth decisioning while staying compliant and auditable.
Skills:
AI Speech-to-Text
AI Agent Development
AI Builder
AI Audio Generation
AI Security
20
126