Public Lab
Varun Innovates
Experiments, prototypes, build notes, and visible limits.
Public lab by Varun Siddaraju
Context-aware spatial systems, notes, and prototypes.
The evidence layer for adaptive XR interfaces and explainable spatial intelligence: public notes, build logs, prototypes, and the limits behind them. This is not the resume or the sales site.
Built by Varun Siddaraju
context-aware systems · spatial intelligence · applied prototypes
50+XR projects and prototypes across education, healthcare, enterprise, architecture, and smart city work.
10+ yearsResearch, product, and delivery experience across XR, spatial computing, and early applied AI work.
2Public book listings: Apress mixed reality plus a self-published systems book.
3Peer-reviewed publications plus research receipts across Sensors, IEEE, and ICDT.
NotesActive research notes across Harmony, adaptive XR interfaces, and multimodal context inference.
Context
Public lab, connected to Varun Siddaraju and VeeRuby.
Varun Siddaraju is the canonical profile. Varun Innovates is the evidence layer. VeeRuby is the commercial delivery company.
Current Artifacts
Public artifacts worth inspecting.
Selected frameworks, interfaces, APIs, and notes that show how the research is being translated into practical context-aware spatial systems.
HarmonyXR_GazePoseContext
A real-time context inference system using gaze, head pose, posture, hands, and spatial context to classify user state as engaged, distracted, transitioning, or idle.
GazeHead PoseHandsBoundaryPosture
OpenSpatialAI API map
An early-stage developer-layer direction exploring APIs, memory, and adaptive behavior patterns for XR workflows. It is active R&D with a limited public surface today.
APIsSpatial MemoryOpenXR
Adaptive XR interface experiments
A rapid-prototyping lane for WebXR, headset experiments, AI-assisted workflow tests, and real-time perception experiments.
WebXRWorkflow TestsPrototype Log
Research Proof Previews
Enough evidence to verify direction, not enough to leak the work.
These are cropped screenshots from in-progress Harmony research materials and a prototype presentation. The PDFs and full presentation are intentionally not published.
HarmonyXR GazePose-Context prototype paper
Draft evidence for the multimodal context-inference prototype. Shown as a cropped title preview only; the paper text stays private while evaluation work continues.
Harmony adaptive spatial intelligence concept paper
Position-paper proof for the Harmony research frame: sensing, context, memory, adaptation, and explainability. Public here as a limited record, not as a full release.
Adaptive XR context prototype presentation
A cropped frame from the research presentation showing the prototype environment. It supports the lab story without exposing the full walkthrough or implementation details.
Public note: these previews show the current research direction without publishing the full drafts or walkthrough. They will stay limited until the work is ready for formal review, submission, or release.
Now + Recent
Current work and recently wrapped threads.
A compact status view of what is active now, what evidence is already public, what was recently cleaned up, and what is planned next.
Active Work
The threads currently shaping the lab story and research direction.
Context-aware behavior inference
Modeling gaze, head pose, hands, posture, boundaries, and task context so XR systems can infer human state instead of reacting blindly.
AI command centers and automations
The build lane is active. Current work is turning the patterns, tradeoffs, and examples into clear public notes.
Product modernization pass
Refactoring an existing product while moving it to a newer version, with the goal of cleaner architecture, compatibility, and release-ready behavior.
Published Evidence
Public work and active research notes that support the lab direction.
Small Teams, Strong Systems
The systems book is publicly listed on Amazon and supports the lab's practical systems-thinking thread.
Harmony adaptive XR research note
The current note clarifies the systems argument across perception, context, memory, adaptation, and explainability.
Site network redesign
Varun Siddaraju, Varun Innovates, and VeeRuby now separate personal credibility, public lab work, and company delivery more clearly.
Done
Recently wrapped or clarified items that explain why the public lab looks the way it does now.
Spatial memory for adaptive systems
The current exploration pass is documented as a foundation for continuity, adaptation, and explainability across XR sessions.
SensAI hackathon mentorship
The mentorship is wrapped. The useful residue is sharper thinking around spatial-context demos, prototype quality, and how teams move from polish to system quality.
OpenSpatialAI and repo cleanup
The repo map now separates public R&D artifacts from docs, internal systems, experiments, and applied XR builds.
Plans
Promising threads that should become cleaner demos, notes, or conversations next.
AI in VR education
Talking with education institutes about integrating AI tutors, adaptive feedback, and richer learning loops into VR education experiences.
Lab Map
One spine: context-aware spatial systems from research to prototype to product.
The map organizes the public lab into connected lanes: research questions, prototype evidence, reusable tools, build logs, product models, and applied lessons.
Harmony questions before artifacts.
The concept paper frames the spine: context sensing, spatial memory, adaptive mediation, and explainability for XR systems that should remain understandable to users.
Harmony concept frame Public preview
Public summary of the sensing, context, memory, adaptation, and explainability argument. Full draft stays private until formal review.
Context model boundaries Active
Defines which signals belong in the model and where the system should avoid pretending it knows user intent.
Evaluation and control Active
Keeps adaptation opt-in, reversible, explainable, and measurable through task outcomes, cognitive load, trust, and well-being.
Prototypes prove behavior, not polish.
The GazePose-Context prototype tests whether gaze, pose, hands, posture, boundaries, and task context can drive adaptive XR behavior without turning the demo into a black box.
GazePose-Context behavior model Prototype
Turns multimodal XR signals into interface decisions that can be logged, inspected, and explained after the session.
Adaptive XR presentation frame Preview
Shows the prototype environment and interaction direction without publishing the full walkthrough or implementation details.
Failure modes Active
Captures where the model misreads attention, posture, boundaries, or task state so the next version is more trustworthy.
Reusable pieces from the lab.
OpenSpatialAI is the developer-facing lane for the same research: context descriptors, spatial memory hooks, adaptive behavior patterns, and evaluation helpers.
OpenSpatialAI API surface Early repo
Maps the developer layer for context descriptors, memory state, adaptive behavior calls, and extension points.
XR instrumentation utilities Active
Collects reusable patterns for gaze, head pose, hands, posture, boundaries, and task-state instrumentation.
Build notes with receipts.
The build log records what moved from private research material into public proof: cropped title-page previews, prototype frames, and clear limits.
Proof preview cleanup Public
Paper previews use cropped title-page regions, and prototype imagery is cropped toward visible content rather than empty frame space.
Artifact promotion rule Active
Strong entries need a screenshot, a tradeoff, a clear outcome, and a boundary on what stays private.
From prototype to useful offering.
Product models translate the Harmony research direction into scoped pilots: adaptive XR learning, command centers, automation support, and VeeRuby delivery shapes.
Adaptive XR education pilots Exploring
Uses context inference and feedback loops to test whether XR learning systems can adapt without overclaiming personalization.
Research-backed services Boundary
Separates public lab thinking from scoped VeeRuby delivery so research proof and commercial claims do not blur.
Lessons from real constraints.
Applied lessons connect the papers and prototype back to field constraints: latency, trust, explainability, session continuity, and user control.
Continuity is not polish Lesson
If systems forget context between sessions, users repeat orientation work. That is a system design problem, not a UI polish issue.
Latency budgets early Lesson
Adaptive XR interfaces need latency thinking across sensing, inference, orchestration, and rendering before demo polish hides the problem.
Items 1-3
Selected Signals
Archive material is only useful when it sharpens the lab story.
Only the signals that strengthen the current research and prototype story belong here.
3 peer-reviewed publications plus active research notes
Sensors, IEEE INFOCOM Workshops, and ICDT give the lab real technical lineage instead of pure personal-brand storytelling.
Best Paper Award, ICDT 2018
This is the only award shown here because it has a clean external receipt and ties directly to the research lineage.
Apress book plus systems book
The Apress mixed reality book is the stronger technical receipt. Small Teams, Strong Systems is a public systems-writing signal with an Amazon listing.
Research, product, and delivery are connected
Texas State research, Ong product work, and VeeRuby delivery are the trail that explains why Harmony exists at all.
Research + Architecture Notes
Harmony is the research frame. OpenSpatialAI is the early developer surface.
Harmony focuses on persistent user models, contextual spatial memory, and explainable AI mediation. OpenSpatialAI explores code/API surfaces for that direction. Both belong here as lab work, not separate company brands.
Harmony
Context-aware spatial frameworkSpatial systems that perceive, remember, reason, and adapt across sessions using persistent context rather than single-session triggers.
Notes
Research notes in motionArchitecture arguments, literature notes, figures, evaluation paths, and submission-oriented research threads.
OpenSpatialAI
Developer layerAn early-stage GitHub project exploring APIs, memory, and adaptive XR behavior patterns. It is intentionally early.
Evidence
Prototype-backed thinkingResearch ideas connect to working demos, code paths, deployment constraints, and model limits.
GitHub Signal
The repo map is broader than the public profile suggests.
The repo map spans public experiments, research systems, documentation, and product-oriented builds with clear boundaries.
Harmony and OpenSpatialAI stack
Open Repository
The research and prototype trail spans HarmonyXR_GazePoseContext, system design work, and adaptive XR interface prototypes for multimodal context inference and practical behavior testing.
Command centers, site systems, and workflow layers
These repos support AI command centers, team workflows, and the sites themselves rather than standalone public libraries.
Writing, docs, and personal system repos
Some repos are not product code. They hold books, documentation, personal operating structures, and knowledge architecture that feed the broader lab.
Delivery, demos, and applied XR product repos
This bucket covers company work, immersive products, vertical demos, and partner-facing builds. Useful signal, but not the same thing as reusable open-source infrastructure.
Demo Library
Selected demo evidence, with the full library available.
The selected tab keeps the page compact. Category filters keep the wider public demo trail available.
LabXR Education
AI-assisted VR education environment that connects earlier XR work with the current AI direction.
AR Periodic Table Game Education
Interactive chemistry learning in AR.
XR Medical Viewer Healthcare
Medical visualization in immersive space.
Human Anatomy — MR Healthcare
Mixed reality anatomy exploration.
Surgical Instrument Testing Healthcare
Testing surgical workflow concepts in XR.
XR City Model Viewer Architecture
Urban-scale spatial review in XR.
Construction Inspection — MR Architecture
Mixed reality support for site inspection and planning.
AR Anchor: Real Estate Architecture
Property visualization with Unity, ARKit, and Azure Spatial Anchors.
VR Welding Simulation Training
Hands-on welding training for Meta Quest 2.
VR Mining Training Training
Immersive training for mining safety procedures.
VR Driving Simulator Training
Virtual reality driving training on Quest.
Multiuser 3D Model Viewer Enterprise
Collaborative engineering design review in AR.
Quest Collaboration Room Enterprise
Virtual collaboration room built for Quest.
PC Holographic Remote Assist Enterprise
Remote holographic assistance on HoloLens 2.
XR Smart City — Clean Energy Smart City
Spatial computing for renewable energy infrastructure planning.
GIS Data Visualization Smart City
3D globe and geospatial data visualization in mixed reality.
Open Full Library
YouTube channel archive
Commercial Translation
Mature experiments move into VeeRuby delivery.
The lab keeps reusable thinking public. VeeRuby carries scoped projects, delivery agreements, and client-facing outcomes.
Boundary
Public lab here. Client delivery there.
Some lab ideas can become client-ready systems through VeeRuby, but only after the user, device, data boundary, success metric, and deployment path are clear.
Here
Varun InnovatesResearch notes, prototypes, SDK experiments, build logs, product thinking, and applied lessons.
There
VeeRubyScoped spatial systems, adaptive XR pilots, training simulations, MR apps, technical consulting, and product builds.
Never
No confidential spilloverClient projects can inspire reusable lessons, but names, constraints, and private details stay out.
Signal
Algorithms over anecdotesPublish the reusable method: adaptive UX logic, rendering patterns, AI workflows, evaluation models, and deployment constraints.
Books
Books that support the lab.
Both books have public listings. The Apress mixed reality book is the stronger technical credential; the systems book supports the operating and builder thread.
Beginning Windows Mixed Reality Programming
Published technical book on hands-on mixed reality implementation.
Open Springer listing
Small Teams, Strong Systems
Published systems thinking book for builders, managers, and founders working under real constraints.
Open Amazon listingContact
For experiments, applied research, collaborations, and serious spatial systems conversations.
Reach out for research collaboration, prototype feedback, talks, or applied spatial systems work. The lab keeps public notes, demos, and references available so the direction is clear before a conversation starts.