I BUILD AI SYSTEMS
THAT WORK BEYOND
the demo.
Agentic AI, RAG, AI reliability, voice systems, backend infrastructure, and cloud deployment — designed and built from prototype to production.
ENGINEERING × intelligence
I build practical AI systems that connect models to real workflows, data, tools, and production infrastructure. My work spans agentic AI, RAG, real-time voice systems, backend services, cloud infrastructure, observability, and reliability — with an emphasis on systems that can actually be tested, deployed, monitored, and improved.
SELECTED AI systems
Four systems built around different layers of the AI engineering stack.
Autonomous Incident Engineer
An autonomous incident investigation system that analyzes production telemetry, investigates failure patterns, and executes structured diagnostic workflows.
Agent Reliability Platform
An open-source observability and reliability platform for tracing AI agent execution, evaluating behavior, and detecting failures such as tool errors, retry loops, latency issues, and reliability regressions.
FROM telemetry_aggregates
GROUP BY 1 HAVING COUNT(*) > 50;
Agentic Data Analyst
An agentic data analysis system that investigates datasets through schema discovery, SQL generation, execution, validation, anomaly investigation, and structured analysis.
AI Voice Employee
A real-time conversational AI system combining audio streaming, speech processing, RAG, tool calling, and agent workflows for customer-facing interactions.
SYSTEM depth
AI ENGINEERING
- ✦ LLM Applications
- ✦ Agentic AI
- ✦ RAG & Vector Search
- ✦ Multi-Agent Systems
- ✦ Voice AI
- ✦ Evaluation & Reliability
BACKEND SYSTEMS
- ✦ Python
- ✦ FastAPI
- ✦ PostgreSQL
- ✦ REST APIs
- ✦ WebSockets
- ✦ Microservices
CLOUD & INFRASTRUCTURE
- ✦ AWS
- ✦ Docker
- ✦ CI/CD
- ✦ Serverless
- ✦ Monitoring
- ✦ Cloud Deployment
PRODUCT & SOLUTIONS
- ✦ Technical Discovery
- ✦ Solution Architecture
- ✦ Prototyping
- ✦ Workflow Automation
- ✦ Product Development
- ✦ Startup / Founder Experience
FROM DISCOVERY TO production
Understand the problem, workflow, users, constraints, and existing systems.
Choose architecture, data flows, models, tools, integrations, and failure boundaries.
Implement the product and infrastructure with clean engineering patterns.
Evaluate behavior, edge cases, failures, security, and reliability metrics.
Move the system into a usable, high-availability production environment.
Measure, debug, iterate, and improve based on real execution traces.