AI infrastructure.Built. Scaled.
I build production AI infrastructure and products — from multi-cloud GPU orchestration and open-source backend systems to evaluation gates, retrieval agents, and multimodal apps.
Outcomes, not adjectives.
AI infrastructure and products.
Production systems, accepted open-source work, measurable products, and public proof where available.
NutrioMN
A Mongolian-first nutrition app using a fine-tuned vision-language model to turn food photos into reliable meal logs.
SkyPilot · Berkeley RISELab
One verified SkyPilot PR merged upstream, five active contributions, and systems research on cross-cloud agent recovery through Berkeley RISELab.
PrimitiveBench
Vendor-neutral evaluation infrastructure for selecting and validating the primitives behind production AI systems.
VAULT Collection OS
A full-stack operating system for tracking, valuing, and understanding high-value physical collections.
CourseLynx
Course discovery agents, catalog ingestion, and Chrome extension infrastructure serving students across 16+ universities.
Curio AI
A LangGraph recommendation system combining live retrieval, structured outputs, and persistent learner progress.
NeuronBook
A grounded PDF learning system with Socratic questioning, concept memory, and a visual knowledge graph.
PM.ai
A multi-agent workflow that turns unstructured project briefs into validated tasks and database-backed execution state.
Google Ads Transparency Monitor
An Apify monitoring primitive that converts public advertising data into repeatable, automation-ready datasets.
I care about the version of AI that survives outside the demo — where real users, real money, and real failure modes are on the line.
I've merged a production infrastructure change into SkyPilot through Berkeley RISELab, with five more contributions active, and built NutrioMN, an angel-funded multimodal nutrition app with 50K+ users, $20K+ MRR, and a #2 App Store ranking in Mongolia.
The throughline is reliable infrastructure around probabilistic systems: scheduling, failure handling, structured outputs, golden datasets, traces, deterministic fallbacks, and CI gates. AI becomes a product only when the system around it is measurable and debuggable.
I move at founder speed but build with engineering restraint — because taste, distribution, and reliability are all part of systems design.
Technical depth, made legible.
Core patterns behind measurable, debuggable agentic systems.
Distributed AI infrastructure
Multi-cloud GPU scheduling, backend services, fault tolerance, and performance.
Evaluation infrastructure
Golden datasets, agent traces, structured validators, and deployment gates.
Production AI products
From multimodal inference to subscriptions, observability, releases, and growth.
Agentic systems
Tool calling, retrieval, ranking, constrained outputs, and external integrations.
Founder speed, engineering discipline.
Production ownership inside small teams and real constraints.
Founder & CTO
Built Dockerized FastAPI services on AWS with PostgreSQL/Supabase authentication, subscriptions, and scan history for an angel-funded nutrition platform, scaling to 50K+ users and $20K+ MRR with a #2 App Store ranking in Mongolia.
Optimized AI/ML inference through prompt compression, caching, and constrained outputs; led 6 interns through Agile sprints, code reviews, monitoring, and production releases while driving 3M+ views and securing 3 gym sponsorships.
Software Engineer
Merged one production PR into SkyPilot main, with five additional contributions in development or review, while developing systems research on cross-cloud inference-state recovery.
Co-founder & Lead Engineer
Built an open-source, vendor-neutral evaluation platform with async execution, multi-tenant job APIs, tracing, and CI/CD gates for production regressions.
Software Engineering Intern
Owned reliable course ingestion for 40,000+ users across 16+ university catalogs, reaching 99.4% reliability and reducing p95 latency from 820 ms to 510 ms.
Software Engineering Intern
Developed APIs, structured recommendation logic, and a LangGraph agent for dynamic content retrieval.
A practical technical range.
Tools for connecting model quality, system reliability, and product delivery.
AI Infrastructure
- Distributed orchestration
- Multi-cloud systems
- GPU workloads
- Kubernetes
- Fault tolerance
- Observability
AI / LLMOps
- Agent orchestration
- RAG evaluation
- Structured outputs
- VLMs
- Vector search
- Evaluation gates
Languages
- Python
- Go
- Java
- C++
- TypeScript
- SQL
Backend / Data
- FastAPI
- Node.js
- PostgreSQL
- Redis
- REST APIs
- ETL pipelines
Product / Infra
- React / Next.js
- Docker
- AWS
- Linux
- GitHub Actions
- CI/CD
Building something that needs to work outside the demo?
Open to AI infrastructure, distributed systems, LLMOps, backend, and high-ownership startup engineering opportunities. Email me directly, or verify the work through LinkedIn, GitHub, and PrimitiveBench.