vaibhav jain
Open to work2 years experienceData Scientist
Data Scientist skilled in LLM evaluation, RAG systems, and time-series analysis.
Standing Rank
Rank Not Available
Developer Badges
No badges earned yet
Skills & Technologies
27 skillsWork Experience
Data Scientist
Turing
- Designed and executed a benchmarking framework using LLM outputs to measure accuracy, robustness, and instruction-following; created a standardized error taxonomy to enable consistent cross-version evaluation and reduce evaluation turnaround time.
- Performed comparative error analysis across model generations to identify systematic failure patterns and improve prompt engineering, shortening iteration cycles between ML engineers and evaluation teams.
- Collaborated with ML engineers to define dataset quality standards and build iterative quality-improvement pipelines, improving dataset quality and producing consistent model performance gains.
Business Analyst Intern
Whiteklay Technologies
- Analyzed and enriched a consolidated banking analytics dataset using AI-assisted workflows and SQL/ETL pipelines, including data ingestion, cleaning, feature creation, validation rules, and metadata capture to improve completeness and schema conformity.
- Defined KPIs for core banking processes and created a structured data catalog covering key business processes, enabling consistent analytics, stronger traceability, and reduced data discrepancies for downstream reporting.
ML Engineer Intern
Credicxo Tech
- Developed CNN-GRU hybrid with autoencoder feature extraction for bearing Remaining Useful Life (RUL) prediction, applying deep-learning techniques to time-series sensor data (CNN, GRU, autoencoder).
- Benchmarked transfer-learning architectures (VGG16, VGG19, MobileNetV2, InceptionV3) for COVID-19 detection from chest X-rays and designed an ensemble model to combine top-performing classifiers.
Projects
Market Regime Detector
- Engineered time-series features from 10 years of Nifty 50 and India VIX data; applied GMM clustering and PCA (6 components) for regime identification, and used Random Forest for validation.
- Validated the detection pipeline with Random Forest test accuracy of 94% and model agreement of 88.9%; RAG-style FAISS retrieval mean confidence ~83.2%.
Real-Time Industry RAG Analyzer
- Designed an end-to-end RAG pipeline that scrapes live articles, chunks documents, generates HuggingFace MiniLM embeddings, and uses FAISS for retrieval to ground LLM responses.
- Delivered a low-latency, source-cited Streamlit chatbot with persistent session memory to reduce stale-knowledge hallucinations, suitable for production deployment.
Leaderboard Standings
Leaderboard Position Pending
Global test scores, peer standing percentiles, and algorithm leaderboard ranks are updated dynamically.
Assessment Highlights
Assessments Not Completed
Coding evaluations, system assessment results, and conceptual score badges will appear here after taking a test.
AI Collaboration Score
AI Collaboration Score Pending
Developer coding behavior, assistant cooperation, and AI pair-programming indicators are evaluated during live coding sessions.
Role Compatibility Profile
Role Compatibility Analysis Pending
Custom matching reports, candidate role compatibility percentiles, and core engineer strength profiles are processed once conceptual code screenings are complete.
Achievements
Achievements Not Earned
Special rewards, developer badges, system recognition certificates, and conceptual milestones will display here.
About Details
Professional Bio
Data Scientist skilled in LLM evaluation, model evaluation, feature engineering, RAG systems, FAISS, embeddings, and time-series analysis. Built benchmarking frameworks and a market-regime detection system, achieving 94% RF test accuracy and 88.9% agreement on 495 unseen trading days.
MBA, Business Intelligence & Analytics; B.Tech, Information Technology
SVKM NMIMS University ( - Present)