Shoxruxmirzo Toyirov
Full-stack engineer with 6+ years of experience building high-performance, large-scale applications across the stack using React, Next.js, Node.js, Python, and modern cloud platforms. Deep hands-on experience with LLMs, AI agents & agentic workflows, LangChain, OpenAI SDK, RAG. Strong background in blockchain, smart contract tooling, and Web3 integrations (ethers.js, wagmi, web3.js). Startup-minded builder who can rapidly prototype, design, and ship end-to-end AI-driven products. Thrives in fast-moving, distributed teams and collaborates closely with AI researchers, auditors, and security engineers.
Experience
Senior Software Engineer
EPAM Systems
Project: CoCounsel (AI-Powered Agentic Workflow for Legal Tech)
- Helped engineering specialized AI agents that autonomously draft, analyze, and format legal documents, reducing manual effort for lawyers.
- Developed the primary product user experience for agentic multi-skill workflow, orchestrating LLMs for case analysis, evidence extraction, and motion generation.
- Developed human-in-the-loop verification layer enabling lawyers to review and approve AI-generated results.
- Built backend components using Python (FastAPI), LangChain, RAG, and frontend features using React.
- Collaborated with AI researchers and auditors to ensure accuracy and security.
Lead Engineer
XCDM Limited
Projects: Kotib AI | TopCoach | Aqcham | Docum AI | Bal Medical
- Led and delivered over 10+ projects in last year alone resulting in over $400k in revenue and improved online presence and credibility.
- Prototyped and shipped AI-driven internal tools, LLM applications, and agent workflows.
- Mentored junior developers, focusing on code quality, performance optimization, and clean architecture principles
- Designed and optimized distributed systems and microservices along with integrating message queue systems like RabbitMQ, ensuring scalability and high availability across services
- Tech: NodeJS, NestJS, NextJS, Express, FastAPI, Django, ReactJS, React Native, Redux, Docker, Google Cloud, Nginx, RabbitMQ, AI/ML (Whisper, OpenAI, & open-source models)
Software Developer
Morrow Digital
Projects: Ownable UK | Moropo | Terra Listens | ACT It Out
- Developed several complex projects including management dashboards and mobile apps, converting Figma designs into pixel perfect prototypes, including rendering 3d designs with ThreeJS (ownable.uk).
- Worked with AWS Device Farm gateways to test and debug mobile apps while implementing clean architecture principles that inherently optimized performance and resource usage, reducing error rates by 20%.
- Implemented serverless architecture using AWS and Netlify
- Built desktop and mobile AI apps in company’s projects and hackathons.
- Tech: NodeJS, ReactJS, React Native, Mobx, AWS, Supabase, Netlify, ThreeJS
Co-founder
Assorty Products
- Architected the entire system from scratch including backend (node.js), frontend (React.js) and mobile (React Native)
- Gained 4k+ users with over 2k orders within a week and scaled the system with high availability
- Improved app performance by 20% using optimized caching, Lazy Loading, and data normalization.
- Set up CI/CD, deployments, push notifications, deep linking which resulted in easier development experience and higher user traction.
Education
Inha University
B.S.
A+ in Programming, Linear Algebra and Databases
Skills
Projects / Startups
Talentere AI
Talentere AI - cursor for recruiters automating the tech recruiting process with AI Agents
- Orchestrate several AI Agents to automate resume screening and candidate interviews including audio calls.
- Developed kanban board style platform for tracking candidates
- Utilized LiveKit platform for realtime audio calls and agents
- Set up a pipeline automating the candidate search, resume screening, conversation with users, realtime calls, and call analysis.
Nexus Stablecoin
Created my own custom Algorithmic StableCoin
- Orchestrate several AI Agents to automate resume screening and candidate interviews including audio calls.
- Developed kanban board style platform for tracking candidates
- Utilized LiveKit platform for realtime audio calls and agents
- Set up a pipeline automating the candidate search, resume screening, conversation with users, realtime calls, and call analysis.
Docum AI
Docum AI is a no-code platform that provides customer service chatbots for businesses using LLMs.
- Won the national startup competition, worth of 50,000 USD
- Built a no-code platform enabling companies to create LLM-powered chatbots
- Designed microservices architecture serving thousands of users monthly.
- Developed semantic caching, RAG pipelines, vector search, rerankers, and integrations (Telegram/Instagram)
- Implemented CI/CD via GitHub Actions & Jenkins.
Languages
English
proficient
Uzbek
native
Korean
limited
Russian
limited