Available for startup projects

Roman Kriuchko

I help early-stage startups ship LLM products, autonomous agents, and RAG pipelines — from prototype to production.

Available for new projects · US timezone · Replies within 24h

Trusted by teams at

Foxquant

Foxquant

Founding AI Engineer

OpenZeppelin

OpenZeppelin

Senior AI & Web3

Aeternity

Aeternity

ML & Blockchain

Geniusee

Geniusee

AI Product Delivery

Intellias

Intellias

Enterprise Engineering

Featured case studies

Computer Agents — Autonomous AI Agent Platform

Lead AI EngineerLangGraph, OpenAI, Anthropic, TypeScript, Next.js

12+

tool integrations per agent

Prod

live SaaS deployment

Multi-LLM

orchestrated model routing

Production agent platform where users deploy autonomous workers for research, coding, and task automation.

View live project →

Foxquant — AI-Driven Quant Trading

Founding AI EngineerLangGraph, PyTorch, RAG, FastAPI, React

0→1

founding AI platform build

RAG

filings & market data retrieval

MLOps

evals · CI/CD · model registry

LLM-assisted strategy research and forecasting models powering a quant trading demo used by traders daily.

View live project →

JSON Scout — LLM Structured Extraction API

AI EngineerOpenAI, FastAPI, eval harnesses, prompt versioning

Schema

validated JSON extraction

Evals

regression & prompt QA

REST

production API surface

Schema-aware JSON extraction API replacing brittle regex pipelines with predictable latency and cost.

View live project →

Nova — AI-Native Web Platform

Full-Stack AI EngineerNext.js, Vercel, LLM workflows, streaming UI

RAG

context-aware responses

Streaming

real-time React UI

Live

Vercel production stack

End-to-end AI product with retrieval over user data and a streaming React experience.

View live project →

What clients say

Roman joined as founding AI engineer and owned our agent stack, RAG layer, and production inference end to end. He ships systems that survive real trading workflows—not demos.

Product LeadershipFoxquant · AI Quant PlatformFoxquant

Strong LLM engineering paired with a security-first mindset. Roman delivered retrieval pipelines and review tooling our team used daily, with clear specs and dependable execution.

Engineering LeadOpenZeppelin · Smart Contract SecurityOpenZeppelin

Roman replaced a fragile regex pipeline with a schema-aware extraction API, eval harness, and predictable cost controls—exactly what we needed to sell to technical buyers.

Technical FounderJSON Scout · Structured Data APIJSON Scout

Services

LLM Product Development

Ship AI-native features: chat, copilots, structured extraction, and workflow automation.

Autonomous Agent Systems

Design agent orchestration, tool routing, memory, and production guardrails.

RAG & Applied ML

Build retrieval pipelines, eval harnesses, and ML integrations with measurable quality.

Full-Stack Delivery

End-to-end ownership from API and models to React/Next.js UI and deployment.

How engagements work

  • Discovery call → scoped proposal
  • 2–4 week prototype sprint
  • Monthly retainer for ongoing AI product work