I am Bharat Singh, Tech Lead for the AI vertical at Milestone Internet and an Information Retrieval (IR) systems architect. (In academic literature, my research is published under B. Singh, and I am known online and on GitHub as Bharat Singh Bhadwal.)
My work centers on the architecture of neural search systems, hybrid retrieval funnels, autonomous agent workflows, and core platform modernizations.
Experience
- Milestone Internet · AI Application Tech Lead (Full-Time) (Sep 2025 – Present): Leading AI platform architecture across autonomous multi-agent systems (LangGraph, FastMCP, MongoDB), multi-tenant RAG across 3,500+ commercial websites (Prefect for offline ETL, local embedders, cross-encoders), and high-throughput CMS rendering engine modernization to Python and Kubernetes.
- Milestone Internet · Enterprise AI Consultant (Contract) (Jun 2025 – Sep 2025): Architected multi-tenant ingestion, DOM scraping, semantic chunking, and vector search infrastructure across commercial web catalogs.
- European Patent Office (EPO) · AI Specialist Consultant (Contract) (2023 – Jan 2025): Architected multilingual information extraction, Word Sense Disambiguation (WSD), and semantic prior-art retrieval pipelines across millions of international patent records.
- Knewco Inc. · Knowledge Graph Consultant (Contract) (2021 – 2023): Developed enterprise knowledge graphs using unsupervised relation extraction and semantic inference engines across large heterogeneous datasets.
- EU COVID Task Force · AI & Data Modeling Lead (Contract) (2020 – 2021): Directed health and mobility data integration and epidemiological predictive ML pipelines across EU member states.
- Medical Informatics Research Consortia / NBIC · AI Consultant & Scientific Researcher (Contract) (2014 – 2020): Engineered semantic web data layers, graph databases (RDF, SPARQL), and clinical text mining engines.
- Erasmus Medical Centre, Rotterdam · Senior Software Engineer & Scientific Researcher (2010 – 2014): Engineered full-stack semantic web architectures, graph stores, and natural language processing pipelines for clinical entity normalization.
- Leiden University Medical Centre / NBIC · Senior Software Engineer (Full-Time) (2008 – 2010): Developed semantic web data layers and unsupervised clinical concept extraction pipelines in Leiden, Netherlands.
- Quark Inc. · Senior Software Engineer (Full-Time) (2004 – 2008): Built enterprise publishing platforms and high-performance document processing engines in Java.
Publications
- The value of data (Nature Genetics, 2011, 43(4), 281–283): Contributed to semantic data modeling and nanopublication specification implementation for machine-readable assertion networks across scientific corpora.
- Using rule-based natural language processing to improve disease normalization in biomedical text (JAMIA, 2013, 20(5), 876–881): Designed and implemented rule-based NLP pipelines and dictionary lookup algorithms for normalizing clinical disease mentions against UMLS and MeSH ontologies.
- Using an ensemble system to improve concept extraction from clinical records (Journal of Biomedical Informatics, 2012, 45(3), 423–428): Developed ensemble concept extraction pipelines combining machine-learning classifiers with rule-based heuristics to extract clinical entities from unstructured records.
Personal Writing
Outside of systems engineering and research, I write personal essays and reflections at heybhanu.com.
Contact & Profiles
- Email: [email protected]
- GitHub: github.com/bsbhadwal
- Resume: View HTML Resume · Download PDF