SOCIAL DISCOVERY · BUILT WITH OVERLIFT

Veluris

Explore a plan and understand why it fits.

Veluris explores event and venue discovery with visible reasons, uncertainty, and consent boundaries. Try the portfolio’s synthetic decision examples to see how a suggestion is evaluated before a person chooses what to do.

Try the Veluris decision examplesOverLift.ai See the evidence
Product system view Explore Veluris, then follow the workflow below.

Recorded project scope

Implemented

.NET and PostgreSQL product foundation with end-to-end owner preparation and testing paths

More in this category (2)
  • Governed adapter and source-reconciliation architecture
  • Shared intelligence fabric covering discovery, planning, venues, social matching, safety, operations, and personalization

Demonstrated

Local ingest and preparation workflow on the owner Mac

More in this category (6)
  • Consent-, safety-, freshness-, and policy-bounded recommendation architecture with Q-Lens path comparison
  • Explainable separation of source facts, learned interpretation, user preference, and official state
  • Offline public-event planning and field-minimized venue enrichment through the R10J local-first laboratory
  • R10M governed-intelligence challenge cases preserving consent, private context, disclosure, contact, and human-choice authority
  • R10K multi-agent world authority across event, venue, preference, safety, consent, explanation, and operations agents without contact or relationship authority
  • R10N contextual product-intelligence laboratory with nine Veluris cases covering current intent, source reconciliation, accessibility, travel, safety, private personalization, clarification, abstention, no-contact/no-send boundaries, receipts, replay, and recovery

Planned

Broader live event, venue, and local news source coverage

More in this category (2)
  • First-user discovery, planning, and introduction acceptance campaigns
  • Production safety, moderation, operations, personalization, abstention, and owner-reviewed recommendation evaluation

Research

Q-Lens social-plan comparison across public events, venues, groups, introductions, safety, accessibility, uncertainty, and abstention without hidden human scoring

More in this category (2)
  • Longitudinal preference learning with explicit expiry and correction
  • Privacy-preserving group, venue, and introduction intelligence
MARKETING

Who it helps and why.

Social products can become invasive or manipulative when inferred preferences are treated as facts, stale events are presented as current, safety context is hidden, or a ranking engine starts acting as though it knows who someone should meet, trust, or contact.

What makes it useful: Many recommendation products optimize engagement around opaque profiles and permanent inferences. Veluris separates observed facts, explicit preferences, temporary context, learned interpretation, safety policy, and consent so a useful recommendation never becomes permission to act on someone's behalf.

BUSINESS / EXECUTIVE

Business value.

Veluris aims to make local discovery and planning easier to explain. The portfolio demonstrates the decision rules with frozen synthetic cases; it does not establish current event availability, relationship compatibility, or real-world safety.

Explore the business outcomes
  • More relevant local discovery
  • Safer and more explainable introductions
  • Personalization without surrendering human agency
ARCHITECTURE

How the system fits together.

Typed plans, fresh-source reconciliation, venue understanding, contextual retrieval, bounded preference learning, specialized safety and uncertainty classifiers, deterministic fusion, and Q-Lens social-plan comparison may propose, rank, explain, personalize, clarify, and abstain. Explicit preferences, rejection history, private notes, drafts, safety context, and plan state stay in a local profile vault by default; remote adapters receive only field-minimized public venue or event fields for an explicit refresh. Current consent, disclosure policy, and typed contact state remain authoritative. They may not contact another person, invent compatibility evidence, infer permanent sensitive facts, override consent, disclose private context, or silently turn discovery into a social action.

A representative workflow

  1. 01

    Choose a planning example and inspect its request, source records, and permitted preference context.

  2. 02

    Compare the options, including reasons that freshness, access, travel, uncertainty, or consent removes an option.

  3. 03

    Read the result and supporting receipt. A suggestion does not authorize contact, disclosure, or a booking.

Explore the architecture details

I designed Veluris around a governed intelligence fabric spanning source acquisition, reconciliation, discovery, planning, venue understanding, social matching, safety, operations, and personalization. Learned components remain proposal systems; consent, policy, provenance, human review, receipts, and replay remain authority.

ENGINEERING

Engineering choices.

I separated source freshness, stated preferences, and current consent before comparing social plans. The portfolio examples show why an option is included or removed and preserve the difference between a useful suggestion and permission to contact someone.

Explore the technology and engineering details
Technology stack.NET 10, PostgreSQL, governed source adapters, local and hybrid semantic search, event and venue intelligence, consent and safety policy, typed recommendation state, receipts and replay
  • One governed intelligence fabric across the complete social product
  • Learned ranking and planning that can explain and abstain
  • Consent, safety, privacy, and human choice as deterministic boundaries
OVERLIFT METHOD

How this product applies the OverLift method.

The OverLift method preserves source freshness, rights, contradictions, exclusions, user-authored preferences, and consent as typed evidence. Agents can compare events, venues, plans, and possible introductions while deterministic authority controls disclosure, contact, approvals, state transitions, receipts, and recovery.

Explore the shared OverLift architecture

The method is reusable. Each product supplies its own facts, rules, execution, and validation. These tiers describe the shared architecture; the project scope above records which capabilities are implemented here.

Why determinism matters here: Human identity, consent, and contact cannot drift because a model inferred a preference or ranked a match. Determinism keeps disclosure, outreach, state transitions, and expiring context tied to explicit authority.

  1. T0
    Exact baseline

    Define the authoritative result before adding acceleration, semantics, or agents.

  2. T1
    Measured evidence

    Instrument state, latency, quality, and failure so improvement is observable.

  3. T2
    Bounded ownership

    Move expensive or high-frequency work behind clear worker and memory boundaries.

  4. T3
    Deterministic transport

    Version messages, snapshots, identities, deadlines, and replay inputs.

  5. T4
    Semantic assistance

    Let local retrieval and models propose meaning only inside admitted evidence.

  6. T5
    Candidate paths

    Compare several allowed next actions with explicit risk, cost, fit, and uncertainty.

  7. T6
    Shadow verification

    Challenge the production route without giving the challenger authority.

  8. T7
    Governed production

    Promote only measured, replayable paths with fallback and operational receipts.

  9. Tier-S
    Experimental challenger

    Run advanced evaluators in shadow, compare them with the control, and graduate only repeatable gains.

CATS

Context-Aware Tier Selection

CATS chooses a qualified execution tier for the evidence, device, latency, permissions, and failure conditions of the task.

SEMANTIC BRIDGE

Meaning becomes typed product state

The Semantic Bridge maps a goal to explicit entities, evidence, uncertainty, tools, and permissions. Missing or contradictory information remains visible.

DETERMINISM

Validate the action and retain the result

Models may retrieve, compare, or propose. Exact software checks facts, permissions, state changes, and execution, then records what happened.

OverLift securityAgentic AI Glossary

Read the OverLift MethodOverLift.ai See other products
PROOF

Evidence and limits.

The linked portfolio lab includes nine Veluris examples and five examples from other projects. It compares frozen synthetic records and shows exclusions, clarification, and receipts. It does not look up live events, identify real people, send messages, or make bookings.

Back to the project overview ↑

Discuss Veluris View resume All projects