Over deze freelancer
🚀 AI Systems & Backend Engineer | Production Ready LLM Solutions
I help companies build AI systems that actually work in production — not just demos.
Most AI projects fail after deployment due to unreliable outputs, scaling issues, or rising costs. My focus is solving exactly that: building robust, scalable, and cost-efficient AI systems that perform reliably in real-world environments.
What I Do
✔ Design and build production-grade LLM systems (OpenAI, Anthropic, RAG pipelines)✔ Develop AI agents and workflows with retries, idempotency, and auditability✔ Improve AI reliability using schema validation, tool contracts, and fallback logic ✔ Optimize performance with hybrid search, caching, and smart chunking ✔ Build scalable backend systems and APIs (FastAPI, Node.js, distributed systems)✔ Ensure data consistency and fault tolerance in high-concurrency environments
What Makes Me Different
Most developers can connect to an AI API. Few can make it reliable.
I specialize in:
-
Preventing duplicate executions, race conditions, and partial failures
-
Designing systems that gracefully handle errors instead of breaking
-
Reducing API costs and rate-limit issues through smart architecture
-
Turning unstable AI outputs into predictable, production-ready systems
Proven Experience
-
Built and operated LLM systems used in customer support, sales, and internal tools
-
Eliminated backend inconsistencies in high-scale systems handling real users
-
Delivered MVPs that successfully scaled into production environments
-
Integrated AI into existing systems without costly rewrites
Example Projects
🔹 AI Customer Support SystemStructured LLM pipelines with real integrations (CRM, ticketing systems)
🔹 AI Booking AgentReliable transaction handling with retries, rollback strategies, and rate-limit control
Tech Stack
Python, Rust, TypeScriptFastAPI, Node.js, NestJSPostgreSQL, Redis, ElasticsearchLangChain, LlamaIndex, OpenAI, AnthropicDocker, CI/CD, Cloud Deployment
Let's Work Together
If you're looking for:
-
A system that won't break among real users
-
AI that produces consistent, usable results
-
A backend that scales without surprises
Send me a message with your project details — I'll help you build it the right way from the start.
Werk & Ervaring
A
10-01-2024 — 17-12-2025
AI Engineer
Exa
- Built and operated production LLM systems used in customer support, sales, and internal workflows
- Stabilized non-deterministic model behavior using schema validation, tool contracts, and fallback logic
- Designed stateful agent workflows with retries, idempotency, and audit logging to prevent duplicate and partial executions
- Improved retrieval performance by implementing hybrid search, optimized chunking, and caching strategies
- Reduced system instability caused by rate limits and cost spikes through adaptive model selection and request deduplication
- Implemented graceful degradation patterns to maintain service under failure conditions
B
20-11-2019 — 13-12-2023
Backend Engineer
ChatPlus Co. Ltd
- Built scalable API and event-driven systems supporting high-concurrency workloads
- Eliminated data inconsistencies caused by race conditions and duplicate events using idempotency and transactional design
- Resolved performance bottlenecks through query optimization, indexing, and asynchronous processing
- Designed zero-downtime deployment workflows with backward-compatible schema migrations
- Improved system reliability using retries, dead-letter queues, and failure isolation patterns
F
19-09-2018 — 11-10-2019
Freelance Full-Stack Engineer
Upwork
- Delivered backend and AI-integrated systems for international clients
- Built MVPs that scaled to production using modular architecture and clear service boundaries
- Integrated AI features into existing systems without full rewrites using adapter patterns and async processing
