.png){kind=logo}
Tapan Patel
Junior Machine Learning Engineer specializing in MLOps and real-time systems
Connect with Tapan
Tapan already has a relationship with Reval, so a warm intro from us gets a much better response than cold outreach.
Recommended
Already have an account?
About
Built and shipped a production GPT-4 + RAG customer support chatbot that materially improved support operations (response time 4 hours to <3 minutes; ~65% tier-1 ticket automation). Demonstrates strong end-to-end LLM engineering across retrieval (Sentence Transformers/Pinecone), safety (multi-layer moderation), cost/latency optimization (caching/streaming, Celery/Redis), and rigorous evaluation/monitoring (shadow deploys, Datadog, 500+ test cases), plus proven stakeholder buy-in leading to 80% adoption.
Experience
Education
Key Strengths
- Deployed GPT-4 RAG support chatbot to production; cut response time from ~4 hours to <3 minutes
- Automated ~65% of tier-1 support tickets without human intervention
- Reduced hallucinations by 80%+ via semantic retrieval + citation requirements + confidence-based escalation
- Reduced API costs by ~40% using aggressive prompt caching and response streaming
- Built multi-layer safety/content moderation (keyword filters + separate classifier model)
- Improved scalability and reliability; reduced P95 latency from 8s to <2s during traffic spikes (Celery/Redis queues, retries, backoff)
- Strong evaluation/observability practice: 500+ test cases, shadow deployments, Datadog monitoring with alerts
- Effective cross-functional collaboration with non-technical stakeholders; drove 80% team adoption through demos and metric-driven iteration
Discover more candidates like Tapan
Search across thousands of pre-screened, high-quality, high-intent candidates on Reval.
Search Talent