Senior engineer with 13+ years building distributed systems, cloud-native platforms, and enterprise software across banking, media, and regulated environments.
My background spans backend engineering, platform modernization, cloud-native architecture, and large-scale operational systems. Today, I focus on applying Agentic AI and AI governance patterns to real-world enterprise workflows where explainability, auditability, and operational reliability matter.
π¦ Software Engineer @ Swedbank
π€ Exploring Agentic AI, Azure AI Foundry, Foundry IQ, A2A and MCP
ποΈ Passionate about Distributed Systems, Platform Engineering and AI Governance
π¬ Banker's WrappedBackblaze Generative Media Hackathon 2026 Spotify Wrapped β but for your bank account. An agentic pipeline that turns a transaction CSV into a personalized 60-second narrated recap video. |
π§ ContinuumCockroachDB Γ AWS Hackathon 2026 Agentic incident-response memory that survives the agent being killed mid-incident β execution state persists in CockroachDB, not process memory, so a cold-started agent resumes from the exact interrupted step. |
π‘οΈ ARGUSMicrosoft Agents League @ AI Skills Fest 2026 Multi-agent compliance intelligence platform on Azure AI Foundry β KYC, AML, and risk investigation through transparent, evidence-backed, auditable decisions. |
| Project | Description | Stack | Status |
|---|---|---|---|
| Banker's Wrapped | π¬ AI-Powered Financial Storytelling β agentic pipeline that turns transaction CSVs into personalized 60s recap videos via NVIDIA NIM, Genblaze, and Backblaze B2 | Python FastAPI Semantic Kernel NVIDIA NIM Genblaze Next.js |
π§ Hackathon WIP |
| Continuum | π§ Agentic incident-response memory that survives the agent being killed mid-incident β execution state persists in CockroachDB for cold-start recovery | Python FastAPI CockroachDB Amazon Bedrock AWS Lambda MCP |
π§ Hackathon WIP |
| ARGUS | π‘οΈ Agentic Risk & Governance Unified Screening β Multi-agent compliance intelligence platform with transparent, auditable, evidence-backed decision making | Azure AI Foundry A2A RAG Semantic Kernel |
π Hack for Good Winner |
| pythonic-algorithms-lab | β‘ CPU/GPU algorithm implementations (sorting, graphs, DP, data structures) with benchmarking infrastructure & Dash dashboard | Python CUDA Dash NumPy |
β Active |
Designing AI systems using Azure AI Foundry, Foundry IQ, multi-agent orchestration, Agent-to-Agent (A2A) communication, RAG with hybrid search, document intelligence, and reasoning-driven workflows β focused on building AI that is explainable, grounded, and production-ready.
13+ years building scalable platforms with Java (Spring Boot, Quarkus) and Python (FastAPI), cloud-native architectures, microservices, NoSQL databases, event-driven systems, and hybrid cloud across AWS, Azure, and OpenShift.
Exploring the infrastructure layer behind modern AI β GPU computing, NVIDIA CUDA, model serving, vector search, observability, and performance engineering.
Designing resilient systems with Azure Functions, Service Bus, OpenTelemetry, Application Insights, KQL, and operational telemetry pipelines that support enterprise-grade reliability and governance.
|
|
|
10+ Professional Certifications including:
- NVIDIA Certified Professional: Agentic AI (NCP-AAI) β NVIDIA π©
- The Microsoft IQ Series: Foundry IQ β Global AI Community
- Machine Learning and AI Foundations: Prediction, Causation, and Statistical Inference β LinkedIn Learning Community
- Kubernetes Administration (LFS458) β The Linux Foundation
- IBM Machine Learning Essentials β IBM
- Python for Data Science β IBM
Building trustworthy AI systems that explain their reasoning, leave an audit trail, and actually work in production.
If that's the kind of problem you're working on β I'd love to talk.