5 years shipping products from idea to production. Now building agentic AI pipelines and reliable systems at scale.
Experience
Tecnoprism
Full Stack Developer (Gen AI) 1.4 years
- Shipped automationcoe.com, an RPA bot builder on 15+ event-driven Node.js services using RabbitMQ, BullMQ, and Kubernetes.
- Built RAG with LLM evals for 100K+ monthly requests; cut MongoDB query latency by 35–40%.
Alois Solutions
Full Stack Developer 2.5 years
- Migrated a large ERP/CRM from Java to Next.js + Express.js on PostgreSQL while keeping production live.
- Optimized queries across 1,000+ tables; virtual scrolling cut render time ~40%.
AI Products
- AI website builder that researches, generates, and deploys full sites from one prompt using agents and hybrid RAG.
- Runs quality evals and background jobs; deploys sites through Cloudflare Workers and R2.
AI Agents Hybrid RAG MCP Microservices RabbitMQ Docker K8s
- 18 specialized agents take startup ideas from validation to deployment, with task-aware RAG and token budgets.
- Selects task-specific context to reduce token usage without losing useful information.
AI Agents RAG LLM Routing Kafka Microservices
- AI video generator that turns a topic into a narrated explainer, with code-as-video scenes built by a multi-agent pipeline.
- Custom audio, secure S3 media storage, and a Cursor-style editor for editing videos with AI.
Python Next.js AWS S3 Docker RabbitMQ
open source
twentyhq/twenty
Open-source CRM · Node.js / TypeScript
- Fixed edge-case null handling in the metadata engine for custom field resolution
novuhq/novu
Notification infrastructure · MongoDB / Next.js
- Contributed MongoDB index optimization for high-volume notification queries
TypeScript ✱ Node.js ✱ React ✱ Next.js ✱ Python ✱ Tailwind ✱ MongoDB ✱ Pinecone ✱ PostgreSQL ✱ Redis ✱ RAG ✱ Applied-AI ✱ Docker ✱ K8S ✱ Vim ✱