01
Trustworthy AI & AI Assurance
- Verifiable agent behavior
- Evidence-backed AI decisions
- Runtime assurance
- AI judges and evaluation
- Bounded autonomy and consequence control
JCEE Labs / Research Portfolio
Research fields
Current research portfolio and interdisciplinary scope across 25 fields, connected by evidence, authority, state, and consequence.
PORTFOLIO REFERENCE · PREPARED
25 fields. One unifying research question.
The current portfolio spans trustworthy AI, distributed and resilient computing, formal verification, quantum information, information theory, causal assurance, complex systems, human-AI interaction, and applied computational intelligence.
UNIFYING RESEARCH QUESTION
How do you determine what a system actually knows, what that evidence permits it to conclude, what authority it possesses, and what consequences it may safely produce?
JCEE Labs operates across multiple research domains, but the work is connected by a common concern: making evidence, authority, state, and consequence explicit enough to test, verify, and audit.
THE CURRENT PORTFOLIO
Research Portfolio
02
Agentic Systems & Autonomous Software
- Multi-agent architectures
- Speculative and branching agent execution
- Agent authority separation
- Recovery and retracing
- Human-agent control boundaries
03
Distributed Systems
- Exactly-once effects
- Crash recovery
- Distributed state
- Cross-host execution
- Coordination failure
- Partial failure and ambiguity
04
Formal Methods & Software Verification
- Executable contracts
- Invariants
- State-machine verification
- Proof obligations
- Reproducible verification gates
05
Authorization, Delegation & Computer Security
- Temporal authority
- Delegation and revocation
- Least authority
- Stale-authority prevention
- TOCTOU safety
- Fail-closed architectures
06
Provenance, Evidence & Auditability
- Evidence graphs
- Cryptographic receipts
- Hash-chained journals
- Provenance tracking
- Reproducibility
- Counsel-grade evidence preservation
07
Causal Inference & Causal Assurance
- Causal direction
- Hidden feedback
- Unobserved backaction
- Observation/intervention separation
- Limits of causal claims under incomplete closure
08
Quantum Information Science
- Quantum information theory
- Entanglement
- Measurement
- BB84-type systems
- Classical/quantum communication limits
- Information-theoretic uncertainty
09
Quantum Foundations
- Measurement and observation
- Uncertainty
- Locality and nonlocality
- Information versus physical state
- Operational interpretations of quantum mechanics
10
Quantum Communication Complexity
- Split-memory problems
- Communication lower bounds
- One-bit communication questions
- Entropic uncertainty proofs
- Distributed quantum-information tasks
11
Quantum Causal Inference
- Unknown feedback channels
- Causal direction in quantum systems
- Measurement backaction
- Hidden variables and context
- Limits on evidential assurance
12
Information Theory
- Entropy
- Conditional entropy
- Mutual information
- Channel capacity
- Communication lower bounds
- Information loss and recoverability
13
Computability / Limits of Knowledge
- What observations permit us to establish
- Epistemic boundaries
- Proof versus inference
- Evidence insufficiency
- Limits imposed by inaccessible state
14
Resilient Computing & Fault Tolerance
- Fault containment
- Recovery semantics
- Ambiguous outcomes
- Dependency failure
- Compensation before collapse
- Coordination reserve
15
Complex Systems & Coordination Science
- Multi-component coordination
- Graph-based state relationships
- Early-warning signals
- Relational precursors to failure
- Emergent system behavior
16
Topology Applied to State Systems
- Orientation-sensitive state
- State equivalence
- Lifted representations
- Context-dependent identity
- Topological analogies such as Klein-bottle work
17
Programming Languages, Runtimes & Compiler Architecture
- Agent/runtime semantics
- Intermediate representations
- Execution contracts
- Compiler/runtime boundaries
- JAIR and LLVM-style ecosystem research
18
Database & Temporal-State Systems
- Bitemporal state
- Versioned authority
- Epochs and generations
- Historical state reconstruction
- Distributed persistence
19
AI Memory & Knowledge Systems
- Retrieval-augmented generation
- Semantic memory
- Vector retrieval
- Persistent personalization
- Controlled cross-application information sharing
20
Human-AI Interaction
- Adaptive AI interfaces
- Higher-order personalization
- Explainable agent behavior
- User control of AI authority
- Trust calibration
21
Machine Learning Evaluation
- LLM-as-judge research
- Deterministic versus learned judges
- Open-weight model evaluation
- Adversarial test fixtures
- Reliability benchmarking
22
Applied AI Safety
- Preventing unauthorized consequences
- Separating hypothesis from action
- Non-broadening authority
- Safe speculative reasoning
- Fail-closed execution
23
Fintech / Payment-System Reliability
- Ambiguous payments
- Capture and payout recovery
- Financial authorization boundaries
- Idempotency
- Reconciliation and evidence
24
Acoustics & Spatial Audio Computing
- Acoustic signatures of spaces
- Frequency response
- Impulse-response-like representations
- Spatial sound characterization
- Creative reuse of environmental acoustics through SIGRESO
25
Computational Culinary Intelligence
- Technique representation
- Flavor and palate modeling
- Substitution reasoning
- Cooking-process sequencing
- Personalized culinary intelligence through MISE
INTERDISCIPLINARY RESEARCH IDENTITY
The work crosses substrates. The boundary stays recognizable.
At the highest level, JCEE Labs sits at the intersection of:
Computer science
Artificial intelligence
Mathematics
Information theory
Quantum information
Distributed systems
Epistemology
THE CONNECTING THREAD
How do you determine what a system actually knows, what that evidence permits it to conclude, what authority it possesses, and what consequences it may safely produce?
This question links JCEE Labs research in trustworthy AI, formal verification, distributed execution, authority systems, causal assurance, quantum information, evidence provenance, resilient computing, and applied intelligence. The individual projects differ in substrate, but they repeatedly test the same boundary: what is justified, what is authorized, and what follows safely from the state actually observed.