Transaction context
Review inputs, model responses, document versions and service calls as related records within a transaction.
SIGILLUM FİNANS
Evidence infrastructure for financial AI
Connect model responses, service calls and version references to each transaction. Review record integrity and coverage gaps within your institution’s own environment.
+
For banks and financial institutions.
Deployed within the institution’s network.
SIGILLUM / THE EVIDENCE LAYER↗
Transaction records.
Cryptographic integrity.[ S / 01 ]
Cryptographic integrity.[ S / 01 ]
Institution-hosted
Cryptographic integrity
Independent verification
Multi-language integration
01 / PRODUCT CAPABILITIES
Transaction context,
integrity and coverage
Record integrity
Verify accepted records through cryptographic signatures, an evidence chain and separate witness records.
Coverage assessment
Compare received records with the events, fields and services defined in your recording scope. Identify missing information.
02 / TRANSACTION REVIEW
ILLUSTRATIVE WORKFLOWReview the records
behind an AI response
Select a workflow to examine the relationship between inputs, reference documents, service calls and recorded outcomes.
TRANSACTION RECORD EXAMPLEIllustrative example
Records linked to one transaction
CONTEXT & REFERENCES↗
Documents and policies used
Review the document, policy and service responses recorded for the transaction, with their version references.
- Document
- Credit assessment policy
- Reference
- Version and content hash
- Record origin
- Service declaration
Retained references show which version was recorded at the time of the decision.
An illustrative example of how it works: review a request, its supporting records, model response and outcome within the institution’s configured recording scope.
ARCHITECTURE VIEW / 01
Within your institution, with separate operating boundaries
From application records to evidence review. The portal and API share an installation; the witness operates in a separate area with its own storage and keys.
THE INSTITUTION’S SERVERS AND NETWORK
APPLICATION LAYER
SDK integration
.NET · Java · Python · Node.js
Transaction, HTTP and message records
TELEMETRY LAYER
OpenTelemetry
OTLP / HTTP JSON
Collector and GenAI mapping
SIGILLUM FİNANS
API and institution portal
Ingestion, coverage and review
Authorized institution access
EVIDENCE STORE
PostgreSQL
Records, signatures and relationships
Encrypted transaction content
SEPARATE OPERATING AREA
Witness service
Recording signed receipts
A separate source for reconciliation
OWNED BY THE WITNESS
Journal and keys
Separate durable storage
Separate signing keys
01
Record delivery
The SDK and collector deliver records to Sigillum through a local durable queue.
02
Witness reconciliation
Signed acceptance receipts and chain information are sent to the witness. Signed witness responses support reconciliation.
03
Institutional control
Production deployment separates the witness’s storage, keys, administrative authority and failure domain.
SDKs AND INTEGRATIONS
View supported integrations.NETJavaPythonNode.jsOpenTelemetryRabbitMQKafka
04 / TEAMS AND RESPONSIBILITIES
A SHARED RECORD OF THE TRANSACTIONFor technology,
risk and audit teams
01
Technology teams
Shared service integration
Integrate shared gateways and AI services once. Connect additional applications through transaction context and configuration.
02
Risk & audit teams
Evidence review
Examine decision records, assess recording gaps and verify exported evidence packages.
03
Infrastructure teams
Institution-hosted deployment
Deploy the portal, evidence store and witness within your institution’s infrastructure and operating boundaries.
TEAM
Banking and technology experience
Our team combines experience in financial systems, enterprise software and machine learning to develop Sigillum Finans for institutional use.
Can Küçükgültekin
Co-founder
Distributed systems · Observability · Applied AI
Can Küçükgültekin works across distributed systems, observability and applied AI. His career includes Akbank, Doğuş Teknoloji and AXA Insurance, with work on API gateway architecture, enterprise logging, Turkish natural language processing and insurance applications. He holds a Computer Engineering degree from İstanbul Ticaret University and completed the University of Virginia’s The Economics of AI certificate program.
LinkedIn
Murat Kırmazel
Co-founder
Banking technology · Payment systems · Systems development
Murat Kırmazel brings experience in banking technology, payment systems and systems development, having held systems development and account management roles at Akbank, Fintek and Intertech. His career also includes enterprise business solutions consulting and team leadership at Metasis Teknoloji, alongside an MBA from Kadir Has University.
LinkedIn
Murat Özer Özaydın
Co-founder
Enterprise architecture · Service integration · Digital banking
Murat Özer Özaydın works on enterprise architecture, service integration and application platforms in banking, with software and architecture roles at Akbank, Intertech and Garanti Technology in his career. He holds a degree in Computer Engineering from Ege University, and his experience includes open banking, digital assistants and enterprise application frameworks.
LinkedIn
Erkan Kıyak
Co-founder
Academic experience · Machine learning
Erkan Kıyak is a faculty member at Pîrî Reis University, with expertise in machine learning methods including decision trees, support vector machines and long short-term memory networks. He holds a PhD in Electronics and Computer Education and a master’s degree in Computer Engineering from Kocaeli University.
LinkedIn