Open to opportunities

Ankit Agarwal

@ankitagarwal4

Message

I build production AI and ML systems for health, fintech, and gaming.

India

Message

What I'm looking for

I'm looking to lead and build production AI and ML systems where I can own work from discovery through deployment, evaluation, monitoring, and stakeholder rollout, particularly in regulated or high-impact product environments.

At Ultrahuman, I lead data scientists and ML engineers building production systems for the Ring, CGM, Blood Vision, PowerPlugs, and health partnerships. I stay hands-on from discovery and system design through deployment, monitoring, and stakeholder rollout.

I led a CX AI platform across the app, WhatsApp, and email at roughly 3,500 daily sessions, combining EKS, Bedrock tool calling, governed knowledge ingestion, hybrid retrieval, reranking, and human escalation. I also built private EKS-based LLM inference for scientific extraction and research workflows.

My health ML work includes Health Age for the Department of Health Abu Dhabi, validated on 28,617 users with 1.45-year MAE, alongside sleep-stage, AFib, sleep-onset, churn, and ring-replacement models. Earlier, I delivered experimentation, lifetime-value, KYC, voice AI, compliance-assistant, risk, and license-optimization systems across PlayGames24x7, WeRize, Salesforce, and EY.

I’ve built regulated models and AI products with rigorous evaluation, release gates, observability, privacy controls, and human handoff. My research includes Blood Age, population-scale sleep disruption, and a patent filing for sickness-onset detection from continuous physiological signals.

Experience

Work history, roles, and key accomplishments

Current

Lead Data Scientist

Current

Ultrahuman

Jul 2026 - Present (2 months)

Leads data scientists and ML engineers across customer experience, sleep, cardiovascular, and metabolic systems, partnering with product, clinical, operations, and external health stakeholders while remaining hands-on in architecture and release decisions.

RAG BM25 Bedrock Pgvector VLLM MLFlow Bi LSTM XGBoost LightGBM SHAP Python SQL PyTorch fastAPI AWS Docker Airflow

Senior Data Scientist

Ultrahuman

May 2025 - Jun 2026 (1 year 1 month)

Led data science initiatives across customer experience, sleep, cardiovascular, and metabolic systems, including building a CX AI Bot and deploying an internal science LLM platform.

RAG BM25 Bedrock Pgvector VLLM MLFlow Bi LSTM XGBoost LightGBM SHAP Python SQL PyTorch fastAPI AWS Docker Airflow

Data Science Specialist

PlayGames24x7

Jul 2024 - Apr 2025 (9 months)

Built survival-analysis and XGBoost lifetime-value models that reallocated about INR 150 million in annual spend toward higher-value cohorts. Ran 20+ powered experiments, contributing to a 12% cumulative lift in net ARPDAU.

Survival Analysis XGBoost A B Testing Python SQL

Senior Data Scientist

WeRize

Oct 2022 - Jul 2024 (1 year 9 months)

Built an internal policy and compliance assistant over PDF sources using PyMuPDF, sentence-transformer embeddings, FAISS, and GPT-4. Deployed a cascaded Hindi-English voice pipeline and built OCR/document classification that reduced KYC turnaround from 3.5 to 1.5 days.

PyMuPDF Sentence Transformers Faiss GPT 4 Whisper BERT OCR Python SQL

Data Analyst

Salesforce

Jul 2021 - Oct 2022 (1 year 3 months)

Built Tableau risk dashboards for about 200 audit and compliance users and used Isolation Forest/DBSCAN on access logs, reducing the measured unauthorized-access risk proxy by about 35%.

Tableau Isolation Forest DBSCAN Python SQL