s
sahill_bharti

Sahil B

@sahill_bharti

AI Product Manager

Inde
Anglais, Hindi
Certaines informations sont présentées en anglais.
À propos de moi
I am a product builder with 5+ years of experience across AI/ML and LLM systems, owning products from strategy to shipment. As a founder of two AI products, I have deep hands-on grounding in RAG and evaluation, allowing me to turn ambiguous problems into sharp product bets and measurable user outcomes.... Plus d’infos

Compétences

s
sahill_bharti
Sahil B
hors ligne • 
Temps de réponse moyen de 1 heure

Voir mes services

Développement de chatbots d'IA
I will build a rag chatbot that answers from your documents with citations
De120 $US / projetPlus de détails
Du site web à l'application
I will convert lovable, base44 or any website to app for ios and android
De100 $US / projetPlus de détails

Portfolio

Expérience professionnelle

advaice

Founder & Product Lead

advaice • Indépendant

Jan 2025 - Present • 1 yr 10 mos

Founded Advaice and took Vikket, a ball-by-ball cricket scoring app, from problem to launch on iOS, Android and web from one React and Capacitor codebase that loads at about 96 KB on low-end phones. I ran the competitive teardown of the category leaders, set a privacy-first stance with no ads and no login, specified an event-sourced scoring engine verified by automated tests, owned App Store and Play Store readiness end to end, and instrumented the acquisition-to-activation funnel in GA4.

Cohere

RAG Data Scientist

Cohere • Temps plein

Jun 2024 - Mar 2025 • 9 mos

Led a team of six with sprint-based workflows and KPIs, driving a 30% productivity increase in two quarters. Integrated RAG into the Command model to lift SQL and Python code-generation accuracy through targeted dataset curation and fine-tuning. Built scalable annotation pipelines for semantic labelling, query disambiguation and multi-hop retrieval, and stood up evaluation frameworks combining automated and human-in-the-loop metrics, raising annotation precision by 15%.

Navisite

Data Scientist

Navisite • Temps plein

Nov 2019 - Sep 2022 • 2 yrs 10 mos

Shipped an AI-driven customer support chatbot that improved engagement and self-service. Built and deployed a support-ticket prioritisation classifier with scikit-learn, NLTK and Azure that cut average response time by 20%, deployed sentiment analysis through Azure DevOps CI/CD with 80% faster integration and testing, and applied churn-risk classification across the client base, helping reduce churn by 20%.