Babita Singh

Healthcare data AI solutions, RAG pipelines, LLM Agents for AI-Powered Drug Development (AIPDD) • I combine 14+ years of health data science expertise with hands-on AI product leadership to turn complex data ecosystems into scalable AI products. • At AstraZeneca, I led the rapid delivery of 11 Generative AI applications built on reusable RAG architectures on AWS Bedrock, transforming fragmented scientific and operational data into AI-ready knowledge systems. • My background in data platforms - from genomics APIs to knowledge graphs and structured analytics, enables me to bridge data governance, engineering, and business strategy. • I excel at translating complex data architectures (SQL, graph, and vector-based systems) into practical AI capabilities. This mix of deep data experience and product execution allows me to industrialize AI solutions that deliver real business value. • I move fast from concept to production, building AI solutions that actually deliver value.

- Advisor for AI-based analytics in healthcare. - Product owner, Project Design - Medical data consulting, Large language models in genomics.

- Core team member of The Beacon project (https://beacon-project.io/) - Product owner; Viral Beacon Project, real-time monitoring of SARS-CoV-2 evolution during Covid-19 pandemic (https://covid19beacon.crg.eu/) - Principal Investigator for La Marató funded pilot project to ‘Beaconize’ hospitals of Catalunya. - Principal Investigator for first EU-funded AI and genomics project Genomed4All: Federated learning & data sharing platform development (https://genomed4all.eu/) - Involved with ELIXIR and GA4GH driver projects to create solutions for a federated (de-centralized) ecosystem for ethical sharing and utilization of human genomics data. (https://elixir-europe.org/news/elixir-and-ga4gh-expand-collaboration)

- Collaborated with the NGO - Made in Carcere, working with women incarcerated in maximum-security prisons. - Integrated science and fashion by designing science-inspired dresses, bags etc. to be sold as NGO products. More info >> https://www.madeincarcere.it/

Whole-genome sequencing data analysis for cancer driver mutations on non-coding regions of the human genome and finding their possible functionality as regulatory regions. Along with some collaborative projects on cancer research.

During my PhD, I was involved in two large scale analysis to understand the regulation of alternative splicing in cancer. Alternative splicing is a mechanism that produces different proteins from same gene sequence and almost 95% of genes in humans undergo this process. Sometime this process produces disease variants of the same protein such as in cancer and other diseases which means the key property of proteins will be lost. We are working with RNAseq data of more than 4000 patients and for more than 12 different tumor types and whole genome sequencing data of >400 patients to produce pan-cancer analysis of splicing changes and its regulation both at cis and trans level.

I visited Bioinformatics group at IMIB hosted by Dr. Konrad Förstner and Jörg Vogel. In my four months stay, I learned advanced python programming (thanks to Konrad's persistence), developed python-based tool for motif enrichment analysis and worked on identifying tumor-specific miRNAs using large-scale genomics data from TCGA.

Data analysis and Database development for stress specific novel miRNAs in Oryza sativa (Rice).

Studied the annotation and mapping of snoRNAs onto the genome of five evolutionary divergent species to understand their localization, specific promoters and their secondary structures formation.

Identification of Environmental carcinogens