Subham Choudhary

Subham Choudhary

+•• ••••• •••••••••••@•••••.com

Data Scientist · Joveo · 4.5 yrs

As a Data Scientist with 4.5 years of experience, I approach complex challenges with diligence and a logical mindset, always striving for mastery in my craft. I thrive on collaboration, leveraging my resourcefulness to uncover actionable insights and drive impactful solutions.

  • Hyderabad
  • Open to Hyderabad · Bangalore · Gurgaon
  • Remote, Hybrid, On-site

Ask about Subham

Skills

Claimed

Strategic ThinkingBusiness AcumenCritical ThinkingRequirements AnalysisAdaptabilityWork Planning and PrioritizationTeam LeadershipMentoring and Peer CoachingAudience Engagement AnalysisKPI AnalysisPerformance Metrics AnalysisAI-Powered Personalization

Proof of Experience

Joveo

Data Scientist

Joveo · Mar 2024 – Present

  • Led development of LLM explanation system and hybrid talent matching platform with explainability, improving matching transparency and increasing recruiter adoption.
  • Owned end-to-end delivery of automated client onboarding system to our talent matching AI platform, and production-grade event-driven services (Kafka + Kubernetes).
  • Reduced onboarding effort from manual workflows to real-time automation.
  • This reduced client onboarding turnaround time from days to near real-time automation.
  • Built unified skill taxonomy system using embeddings + LLM aliasing, improving skill normalization accuracy and reducing duplicate skill mappings by 40%.
  • Developed multilingual document classifier integrated into parsing pipelines, improving document classification precision to 95%+ across resumes and job descriptions.
  • Designed location intelligence platform using Google APIs with spatial analytics, improving geo-search relevance and reducing search latency by 35%.
  • Optimized core services with caching, pagination, and filtering strategies, reducing API response times by 45% in multi-region deployments.
Hood

Data Scientist

Hood · May 2022 – Feb 2024

  • Built recommendation systems and LLM-based pipelines that improved user engagement by 15%.
  • Developed text classification & content moderation models deployed in production, reducing spam and unsafe content exposure significantly.
  • Applied RAG with LLMs to suggest comments, yielding 7% engagement increase for existing users and 25% for new users.
  • Designed scalable ETL pipelines and optimized ML models for low-latency inference, reducing model serving latency by 50%.

Proof-of-Work

No completed sessions yet. Anything Subham finishes shows up here with the evidence behind it.

Proof of Education

Degree, B.Tech

NIT Rourkela · 2017 — 2021

A Proof CV — one profile, kept current, with the proof attached. Back to the portal