Ghasem Nobari, PhDDirector · AI & Computational Sciences
Director · AI & Computational Sciences
Innovation, all the way to validated production.
Research→Ideation→Innovation→Prototype ↻→Validated production
I build and lead teams that solve complex AI problems. I stay hands-on from rapid prototyping and tool selection to architecture and production, while pushing into agentic systems and paths toward AGI.
How I turn ideas into validated systems, across Novartis, Optum, Nokia Bell Labs, Leadbook and NUS. Each diagram shows, step by step, how a real system works: from ideation, through prototype loops, to production.
Data modalities, over time
From language and vision to molecular systems.
Across five career eras, the data changed with the problem. The through-line stayed the same: turn difficult evidence into a system people can use.
12modality families across 5 eras
012008-2016
Language becomes structure
Research in language, graphs, vision and temporal data became commercial entity intelligence.
Director, AI & Computational Sciences: taking AI from ideation through prototype loops to validated production across discovery, R&D and enterprise; collaborations with the NVIDIA, Microsoft, OpenAI and AWS teams.
De novo · design
MD simulation · 80 GPU→Trajectories → features→Fine-tune ML emulators→Benchmark vs lab → repeat
01 · Novartis
AI-Driven Drug Discovery · Galaxy Award ’25
Biologics AI Platform
An integrated antibody-discovery platform built around a reinforcement loop. Large-scale molecular-dynamics simulations are orchestrated across ~80 GPUs on DGX; their trajectories feed feature extraction that fine-tunes ML emulators of biophysical and developability properties: stability, aggregation and manufacturability. Predictions are benchmarked against wet-lab results (lab-in-the-loop), and the gap drives the next round. The emulators cut compute cost while preserving accuracy, alongside de novo design and binding-affinity optimisation.
≈ 40%faster screening
~80GPU MD orchestration
2025Novartis Galaxy Award
Molecular dynamicsML emulatorsReinforcement loopGenerative de novo designAffinity optimisationNVIDIA DGX Cloud
An end-to-end multimodal platform that automates medical-claim and material validation. The hard part was the inputs: messy, unstructured lab reports with embedded charts and tables. Custom-built models parse those charts and tables into structured fields, a library of ~20 specialist models is orchestrated over them, and a unified evaluation layer with human-in-the-loop gates and full audit trails turns a multi-week manual review into a near-real-time, traceable workflow.
PK assay reports hide their numbers in some of the hardest tables in pharma: multi-page, side-by-side, paragraph-shaped, with grouped keys (Fu mean / SD) and units that sit in a different cell. I reframed extraction as a spatial search problem. A custom layout model finds the tables and columns, then a graph method walks the natural reading order to pair every key with its value. It is config-driven and unsupervised, so scientists define new extractions themselves: no labelled data, a confidence score on every page, and under a second per page on a laptop. It outperformed heavyweight commercial and deep-learning extractors that needed thousands of labels and still missed complex layouts.
< 1sper page · on a laptop
0 labelsunsupervised · config-driven
2023Novartis Galaxy Award
Custom layout modelSpatial / graph pairingSME self-service configGrouped keys + unit linkingConfidence per pageHuman-in-the-loop QA
Ideate + build the library→Experiment + benchmark on GPU/cloud→Report · loop the cycle
04 · Novartis
Agentic · Autonomous R&D
Multi-Agent R&D
A first-of-its-kind agentic system for continuous, autonomous R&D, running as an iterative loop: ideation, then agents that build their own library of papers and tools end-to-end, exposed through a library agent over tools and MCP that both other agents and human researchers can query. Specialist personas for orchestration, planning, search, code, critique and writing run across 10+ heterogeneous LLMs, with shared memory, human checkpoints, GPU/cloud orchestration, in-silico benchmarking and a comprehensive report after every cycle.
AI-driven therapeutic-target identification that integrates multi-omics and high-dimensional biology with graph-based learning to surface and rank novel targets. A companion biomarker-discovery effort (imaging + proteomics) advances predictive signatures across neuroimmunology programmes, including progressive multiple sclerosis.
Retrieve from sequences → labels → literature→Extract, then critic agents challenge→Validate with traceable provenance
06 · Novartis
Agentic AI · Immunogenicity
Immunogenicity Evidence Extraction
A multi-agent RAG system that mines and validates immunogenicity evidence from VH/VL sequence-level signals all the way to clinical findings, FDA approvals and labels, patents and published studies. Extractor agents pull candidate evidence; critic agents challenge and cross-check it; a validation step enforces structured, dataset-level QA with traceable provenance, feeding downstream developability and risk assessment with evidence you can audit back to source.
Brief + brand & compliance rules→Generate with an LLM→On-brand copy · layout · visuals
07 · Novartis
Generative AI · Marketing
Horizon-X
A sibling to Horizon MAP that turns the same governed foundation toward creation: an LLM system that generates marketing and communication materials (copy, layout and on-brand visuals) from a short brief and a set of brand and compliance rules, with review built in. The same orchestration and guardrails that validate documents are reused to produce them.
More from this eraMicrosoft AI Empowerment 2021Responsible-AI governance EU AI Act
Optum · UnitedHealth2018-2020
Principal AI/ML Scientist, AI Innovation (iLab): production AI for affordability, payment integrity and clinical decision support inside a regulated US health plan.
Claims → spend driver
Embed claims with a custom model→Track drift across space · time · layers→Surface the spend driver
08 · Optum
Payment Integrity · Custom Embeddings
Bi²: Behaviour Signal Intelligence
A system that transforms raw claim data into temporal signals using a custom-built embedding model, then tracks how those embeddings move through space and time across different layers of the healthcare system (facility, provider and member). Where an embedding drifts away from its neighbours, a previously invisible medical-spend driver surfaces, turning claims into prioritised, explainable affordability levers with dynamic dashboards to explore them.
An engine that automates medical cost-saving ideation. It reads clinical and claims signal across the population and translates it into a ranked set of actionable, explainable affordability levers, surfacing where care can be delivered better and cheaper and handing analysts a prioritised worklist instead of a blank page.
1US patent
rankedactionable levers
explainableanalyst worklist
Clinical + claims signalCost-saving ideationRanking / prioritisationExplainable AI
Claim → pay / review + why
NLP reads the claim→XAI explains the call→Recommend · pay or review
10 · Optum
Payment Integrity · NLP + XAI
AutoD: Automatic Payment Decision Helper
A high-throughput assistant for claim review and payment integrity. NLP reads each claim, an explainable-AI layer surfaces the reasons behind every call, and the system recommends pay-or-review under strict audit and compliance constraints, keeping a human in control while clearing the routine volume fast.
NLP + XAIexplainable review
audit-gradecompliance constraints
HITLhuman stays in control
NLP / NLUExplainable AI (XAI)Payment integrityAudit & compliance
More from this eraUnsupervised fraud & anomaly detection claims-scaleProvider behavior monitoring self-supervisedDynamic dashboards D3 · Plotly
Nokia Bell Labs2016-2018
Lead AI/ML Scientist, Innovation & AI: multimodal AI, large-scale media systems, and human-sensory interfaces (EEG / ECG / EMG / eye-tracking).
Text · image · video → one index
0modalities0unified index
Ingest text · image · video→Multimodal classification→Unified index + search
11 · Nokia Bell Labs
Multimodal AI · IEEE
Smart News Aggregation
A large-scale smart news-aggregation system built in the lab: it ingests news as text, image and video, classifies it with multimodal models, and organises everything under shared, machine-generated topic labels so the whole corpus becomes searchable and linkable across formats. The indexing core (ANNOTATE) was demonstrated at Mobile World Congress 2018 and published at IEEE Big Data.
Read gaze on cycling images→Descend the ImageNet hierarchy→Confidence > 80% → top-5
12 · Nokia Bell Labs
Temporal Signal AI · Biosignals
Temporal Biosignal Models
Temporal models for passive human-computer interaction from EEG, ECG, EMG and eye-tracking signals. The flagship demo infers intent from gaze: you think of an object and stand before a gaze-tracking screen; as images cycle, the model reads where your eyes settle and walks down the ImageNet hierarchy, narrowing from broad categories to specifics until confidence passes 80% and it returns its top-5 guesses. No clicks, no typing; intent inferred from gaze.
A scalable pipeline that turns raw video into structured, multimodal object chunks. It ingests media at scale, segments each video, and runs fast deep-learning models to label every segment, finding the inner similarities and relations across a library so that video becomes searchable, linkable content rather than an opaque stream.
scalableoptimised ingestion
fast DLsegment labelling
linkedinner similarities
Video segmentationDeep learningMedia ingestionSimilarity / relationsSpark
Message → personalised
Infer meaning · hierarchical topics→Compact, transferable representation→Personalised re-expression
14 · Nokia Bell Labs
NLP · Personalised Meaning
Smart Communication: Meaning Transformation
A system to represent, infer and communicate meaning for personalised content. Unsupervised hierarchical topic models, with experiments in deep generative networks, learn a compact representation of meaning that can be transferred and re-expressed per user, with evolving disambiguation of senses and generalised topic labelling across multimedia, so the same message adapts to each recipient.
More from this eraIndoor navigation & IoT context 2 patentsEEG / EMG passive interaction UX research
Leadbook2014-2016 · Singapore
Senior AI/ML Scientist & R&D Lead: built one of Asia’s largest B2B intelligence platforms, end to end from data engineering to learned recommendation.
B2B graph · recommender
Crawl & merge millions of records→Model company–product–customer ties→Recommend the right prospects
15 · Leadbook
B2B Intelligence · 11 Patents
B2B Graph & Prospect Recommender
One of Asia’s largest B2B intelligence graphs, built from tens of millions of verified company and contact records merged from across the web. Its prospect recommender uses a patented Company-Product-Customer “deep relationship” model to learn which new prospects resemble a customer’s best existing ones. The engineering ran from distributed crawling and entity-matching to real-time lookup.
44M+verified contacts
11.5Mcompanies
11Singapore patents
Deep Relationship ModelSpark · HadoopElasticsearch (custom plugin)Go proxyReact extensionTorch
More from this eraReal-time lookup proxy GoContact-lookup browser extension React.jsPersona-scoring plugin ElasticsearchFeature classifier TorchEmail-validation service
NUS & Singapore2008-2014
PhD & Researcher, School of Computing, National University of Singapore (A*STAR SINGA), #1 in Asia for Computer Science (QS); NLP, topic models and privacy-preserving data systems, alongside ventures and AI-for-good prototypes.
Discussions → aspect / action topics
pricing · complaint.92
feature · request.85
support · praise.77
delivery · issue.69
UX · suggestion.61
Millions of flat posts→Joint aspect–action topic model→Temporal aspect/action graph
16 · NUS (PhD)
NLP · AAAI
Aspect–Action Discussion Graph
Doctoral work that takes millions of flat user comments and posts and turns them into a temporal aspect-action graph: a joint aspect-action topic model infers what people are talking about and what they intend to do without labelled data, then arranges it into a structured, time-aware hierarchy of who said what about which aspect, when. Published at AAAI; the foundation of a self-supervised discussion-analysis and prediction system, supervised by Prof. Chua Tat-Seng. Thesis: Learning and Modeling the Underlying Semantics of Online Discussions, ScholarBank@NUS.
More from this eraClinical anonymization: PASS DASFAA 2010Maritime geolocation graph · topic modelsVisual model for the blind GTC 2015
Independent product labs
Outside enterprise work, I prototype product ideas to stay close to users, markets and emerging AI interaction patterns.
intake
review
strategy
deposition
radar
privilege ✓
a matter → six modules → defensible output · audit trail
AI · Legal Platform
The AI Cockpit
A comprehensive AI platform for legal teams: six connected modules that carry a matter from intake to defensible output, with cited reasoning and an audit trail running through all of it.
Smart Inbox & Matter TriageReads, routes and opens matters from the inbox
Document ReviewReviews documents and answers with citations
Strategy WorkbenchBuilds and pressure-tests case strategy
Deposition ArchitectPlans depositions, reconstructs the evidence timeline
Clients & Regulatory RadarTracks clients and shifting regulation
Privilege SentinelGuards privilege and keeps the audit trail
A physics-based, pixel-level OLED guardian that protects against burn-in and keeps working inside games by automatically detecting static HUD elements and treating them in real time.
A VR chemistry-lab training simulation, prototyped with a custom liquid-interaction engine and a library of lab protocols so learners can pour, mix and run procedures with believable fluids in a safe virtual lab.
A personal operating system of AI agents that helps you run your life by capturing, planning and offloading the mental tasks you'd otherwise juggle, so more of the day's overhead runs itself.
A system that helps you get 1% better across the parts of life you care about through small, AI-guided adjustments that compound into real change over time.
Habit trackingAI nudgesCompounding goals
0steps
directed steps · follow · fork · improve
Knowledge · Processes
Global Process System
A “Wikipedia of processes”: one place gathering the steps for anything with a procedure, from applying for a job or visa to onboarding at a company or becoming an Olympic medallist. Each process can be followed, forked and improved.
Structured processesVersioningCommunitySearch
Accessibility · Wearable
Sonar Ring
A wearable smart ring that fuses on-device AI vision with ultrasonic / sonar ranging to perceive obstacles and open space, guiding blind and low-vision users with real-time directional haptics. Prototyped in Singapore.
Edge AIComputer visionUltrasonic sensingHaptics
Venture · Startup Weekend
MysteryShopper
A crowd-sourced, location-based mystery-shopping platform that matches tasks to the right people and places using contextual topic models, turning everyday shoppers into a distributed sensing network. Winner, Startup Weekend Singapore.
Topic modelsGeo-matchingCrowdsourcingMobile
Health AI · Vision (GTC ’15)
Health-Aware Shelf Scanner
Point a phone at a supermarket shelf and a vision-and-health model recognises each product and paints a personalised health “hue” over it. Runs on the first structured database of Singapore food labels, which we built and processed to train it.
