From complex workflows
to dependable software.
I build the APIs, interfaces, and AI integrations behind products people use. From conversational agents to operations platforms, I focus on reliable systems, faster workflows, and lower running costs.
70–90 → 6–9
Daily AI spend at Botsify
Routing workloads across five providers
100 → ~10 / month
MCP integration costs
Migrated from Pipedream to Smithery
Backend to production
End-to-end ownership
APIs, queues, databases, and infrastructure
Selected work
Real products. Hands-on engineering.
Explore what I built, the problems I worked through, and the engineering behind each product.
Drag to explore
Botsify Agentic
AI automation with lower costs and resilient provider routing
Pharma Operations
Connected inventory, manufacturing, and sales
Realmex
Property discovery that keeps agents and clients in sync
Cabbie
Ride operations with live driver tracking
GigsFinder
From venue discovery to concert bookings
The stack behind the work
Laravel and NestJS for the backend. Vue and Next.js for the interface. Cloud infrastructure and AI integrations to bring it all together.
Languages & Frameworks
APIs, background services, and responsive product interfaces
Databases & Storage
Data modeling, query optimization, and caching
DevOps & Cloud
Production deployments, migrations, and infrastructure management
AI & Integrations
Agent workflows, tool integrations, and provider routing
What I bring to your product
Hands-on engineering across the interface, backend, and infrastructure, grounded in the needs of the business.
AI that fits the workflow
Conversational agents, MCP integrations, and web and file search that connect AI to the tools and information a product needs.
Interfaces for everyday work
Dashboards, admin panels, and real-time views that help teams manage inventory, conversations, bookings, and operations.
Performance with a practical budget
Provider routing, database tuning, queues, and caching. At Botsify, I reduced AI spend and integration costs while simplifying production infrastructure.
Beyond the first release
I work across APIs, background jobs, databases, and deployment, and investigate production issues when real usage exposes the edge cases.
Experience across business domains
From pharmaceutical operations and property alerts to taxi tracking and live music bookings, I turn business rules into usable product workflows.
Resilience built into the system
LLM provider fallbacks, isolated MCP sessions, booking safeguards, and support tools that help teams recover when things go wrong.
Notes from the engineering work
Practical deep dives into AI infrastructure, agent integrations, and the details that shape production behavior.
Understand KV Cache Step by Step
A hands-on walk through KV caching in LLMs: token caching, eviction, LRU, paged blocks, prefix reuse, preemption, scheduling, and a unified engine.
Read articleAre LLMs a Single Point of Failure for Your Business?
Why cost-aware, hybrid AI architecture beats wrapper apps, plus practical fallback, routing, and caching infrastructure patterns.
Read articleUnderstanding OpenAI json_schema, MCP, and the 30-Second Streaming Pause
Why missing tool_choice when using OpenAI structured outputs with MCP causes a long streaming pause, and how to fix it.
Read article