Founding Engineer building AI agents, real-time systems, mobile infrastructure, and 0→1 products.

I’ve spent the last 4+ years working inside early-stage startups, usually at the messy stage where the product is unclear, the system is not mature yet, and someone has to figure things out end to end.

Currently at Huddle01, building high-throughput Web3 infrastructure and real-time systems. Previously at Nintee, an AI-native healthcare startup, and GeekyAnts, where I worked across React, React Native, and full-stack product engineering.

I like problems where ownership matters more than job titles.

I ship. I own. I learn fast.

• Building 0→1 products from vague ideas
• Taking ownership across frontend, backend, mobile, infra, and product
• React Native, native modules, mobile performance, and real-time UX
• AI agents, context engineering, tool orchestration, and agent workflows
• High-throughput event processing, queues, caching, workers, and APIs
• Moving fast in small teams without needing perfect requirements
• Learning unfamiliar domains quickly and turning them into shipped products

2024 – Present

Building real-time communication products and high-throughput Web3 infrastructure.

At Huddle01, I’ve worked across mobile, native systems, real-time audio/video, and backend infrastructure. I helped build and scale React Native apps with custom native modules, WebRTC flows, audio/video experiences, and performance-sensitive mobile UX.

More recently, I’ve been building a high-throughput trading platform designed for low-latency Solana/Web3 data processing. This includes parsers, background workers, Redis caching, token ranking, APIs, microservices, and systems that process large volumes of real-time blockchain activity.

Areas: React Native, Expo, WebRTC, native iOS/Android, Node.js, Bun, Redis, ClickHouse, Postgres, Solana, real-time systems.

2023 – 2024

AI-native healthcare startup.

Worked in a small team searching for product-market fit and shipped multiple production-grade prototypes across iOS, Android, and Web. Built early AI-native experiences before AI coding agents became mainstream, including RAG-based flows, agentic product experiments, and a Gen-UI style system that could turn prompts into structured UI using predefined components and rules.

This is where I became deeply interested in AI-native consumer products, especially healthcare, personalization, and products that feel useful beyond a basic chatbot.

Areas: React Native, React, AI-native UX, RAG, agentic flows, rapid prototyping, product iteration.

2021 – 2023

Worked across React, React Native, and full-stack product engineering.

Built and shipped production mobile and web applications for client projects across different domains. This gave me a strong base in product engineering, cross-platform development, frontend architecture, and working with real production constraints.

Areas: React, React Native, JavaScript, TypeScript, Redux, Node.js, mobile apps, web apps.

An experimental Android control layer that gives AI agents the ability to operate a real phone through commands like tap, swipe, type, read UI tree, and navigate apps over WebSocket.

Built to explore a simple question:

What happens when an AI agent can actually use a phone instead of only generating text?

Repo: https://github.com/ponikar/phone-for-ai-agent

Developer tooling experiment focused on improving engineering workflows with AI.

Repo: https://github.com/ponikar/gitfix Live: https://gitfix.ponikar.com

An AI distribution agent experiment.

The idea is to help products find distribution by researching communities, generating content, testing channels, learning from results, and reporting what worked.

Live: https://attentionspam.ponikar.com

Experiment around running local/on-device AI models inside React Native.

Repo: https://github.com/ponikar/react-native-local-model

React Native native library for detecting connected audio devices, including wired headphones and Bluetooth devices, with TypeScript support.

Repo: https://github.com/ponikar/react-native-headphone-detect-v2

Experiments around React Native new architecture, JSI, Turbo Modules, video, device APIs, and native performance.

Repos:

• https://github.com/ponikar/react-native-turbo-video
• https://github.com/ponikar/react-native-super-turbo-device-info

Languages: TypeScript · JavaScript · Go · Kotlin · Swift · Objective-C basics

Frontend: React · React Native · Expo · Next.js · Tailwind · Redux · Zustand · React Query

Mobile: React Native New Architecture · Native Modules · JSI · Turbo Modules · Expo Modules · WebRTC · Reanimated · MMKV · NativeWind

Backend: Node.js · Bun · Go · Express · Hono · REST APIs · WebSockets · background workers

AI: AI agents · context engineering · RAG · tool calling · agent workflows · local model experiments · prompt-to-action systems

Infra/Data: Docker · Kubernetes · Redis · ClickHouse · Postgres · MySQL · MongoDB · Firebase · AWS EC2 · S3 · CloudFront · Nginx · PM2

Web3: Solana indexing · Helius RPC · token lifecycle tracking · real-time trade processing · high-throughput event pipelines

I’m looking for Founding Engineer / Early Engineer roles at AI-native startups.

The best fit is a small, ambitious team where I can own large parts of the product and system, work close to users, and help turn ambiguous ideas into shipped products.

Focus areas:

• AI agents
• Agent infrastructure
• Context engineering
• LLM product systems
• Developer tools
• AI-native consumer products
• High-throughput infrastructure
• Real-time systems

Open to remote roles, US/EU teams, early-stage startups, and meaningful equity.

I care about speed, ownership, and useful products.

I’m not attached to one layer of the stack. If the biggest bottleneck is frontend, I’ll fix frontend. If it’s infra, I’ll go into infra. If it’s product clarity, I’ll talk to users. If it’s distribution, I’ll try to understand the market.

I’m a generalist by nature and a specialist when the problem demands it.

Last updated: July 2026