Гайда

DANYLO HAIDA
FULL-STACK AI ENGINEER | AI PRODUCT & LLM SYSTEMS
Uzhhorod, Ukraine | [open contact info](look above in the "contact info" section) | [open contact info](look above in the "contact info" section)
[open contact info](look above in the "contact info" section) | github.com/HaidaDaniel

PROFESSIONAL SUMMARY
Software engineer with 2+ years of commercial React/TypeScript experience and hands-on applied AI engineering
experience. Built independent systems for RAG and hybrid retrieval, durable agent memory, LLM orchestration, MCP
tooling, evaluation, and local/self-hosted model workflows. Strong product frontend background with backend
experience in Node.js, Python/FastAPI, PostgreSQL, Redis, APIs, and Docker.

TECHNICAL SKILLS
AI / LLM: RAG, hybrid retrieval, embeddings, reranking, vector search, context engineering, durable memory, MCP,
agent tooling, LLM evaluation, provenance, local LLM workflows
Backend: Python 3.12, FastAPI, Pydantic, Node.js, TypeScript, Express, REST/OpenAPI, PostgreSQL, SQLite,
MongoDB, Redis, Prisma
Frontend: React, Next.js, TypeScript, TanStack Query/Router, Redux Toolkit, Zustand, React Hook Form, Tailwind
CSS, shadcn/ui, Radix UI, Ant Design
AI Infrastructure & Quality: Qdrant, Docker, Linux, OpenAI-compatible APIs, llama.cpp, vLLM, pytest, Vitest/Jest,
smoke tests, golden/eval fixtures, quality gates

AI ENGINEERING PROJECTS
InfoWarden — Self-hosted Context, Retrieval & Durable Memory for AI Agents | 2026
github.com/HaidaDaniel/infowarden | TypeScript · Node.js · PostgreSQL · Redis · Qdrant · Prisma · Docker · MCP
• Designed isolated workspaces with immutable revisions, deterministic parsing/chunking, incremental refresh,
provenance, and hard context budgets.
• Implemented PostgreSQL lexical retrieval, Qdrant dense retrieval, optional hybrid ranking/reranking, durable
memory, REST APIs, Studio UI, read-only MCP tools, and provider-backed cited RAG answers.
• Built a reproducible local demo and reliability/evaluation tooling with embedding-provider validation, smoke
coverage, golden retrieval benchmarks, and rebuildable vector indexes.
Novel Harness — Resumable Multi-Model LLM Translation Platform | 2026
github.com/HaidaDaniel/novel-harness | Python 3.12 · FastAPI · Pydantic · SQLite · OpenAI-compatible APIs · MCP
• Built a resumable Japanese-to-Russian LLM translation platform with persisted runs, pause/resume recovery,
local/API model profiles, sample-first approval, human review/final layers, and EPUB export.
• Exposed the same persisted workflow through FastAPI, Web UI, CLI/TUI, and typed MCP tools while preserving
source, prompt, model, decoding, and usage provenance.
• Published a benchmark across 10 model/configuration variants and 100 saved candidate translations using
anonymized AI evaluation and recorded usage/cost evidence.

AI ENGINEERING PROJECTS (CONTINUED)
AgenticProjectKit — Repository-Local Workflow CLI for AI-Assisted Development | 2026
github.com/HaidaDaniel/AgenticProjectKit | TypeScript · Node.js · CLI tooling · CI / quality gates
• Created an open-source CLI for structured AI-assisted development with bounded task contracts, dependency
tracking, allowed-path scopes, acceptance criteria, candidate-bound verification, and review evidence.
• Implemented repository context generation, completion gates, freshness checks, brownfield adoption, planning
skills, release validation, and automated coverage gates (90% lines/statements, 95% functions).

PROFESSIONAL EXPERIENCE
AppFlame — Front-end Developer | Macro Minder, Telegram Mini App | Jul 2025 – Oct 2025
• Owned the frontend scope for onboarding, food logging, dashboard, profile, and settings in a production Telegram
Mini App.
• Integrated AI-powered food logging via photo analysis into a unified daily tracking flow and iterated onboarding
based on user drop-off data.
• Built product visualizations and integrations with React, TypeScript, TanStack Query/Router, shadcn/ui, Radix UI,
Recharts, Framer Motion, and Google Maps/Places APIs.
Axels — Front-end Developer | Outstaffing | Sep 2023 – Jun 2025
Enterprise recruitment platform · Telemedicine SaaS · Shopify application · Backend integration projects
• Built and maintained complex React/TypeScript interfaces including 40+ column data tables, server-side
pagination/filtering, multi-step forms, role-based workflows, dashboards, scheduling, and typed API integration.
• Migrated a production Next.js 13 application to Next.js 14, resolved Ant Design build issues, and maintained
Jest/React Testing Library coverage with an 80%+ CI threshold.
• Built and published a Shopify app from scratch and delivered an Express/TypeScript MongoDB-to-PostgreSQL
synchronization service with feature flags and incremental updates.
Confidential Fintech — Front-end Developer | NDA | Mar 2023 – Aug 2023
• Extended a microloan SaaS product constructor and document editor, fixed card-verification and legacy React
issues, and delivered a course aggregator MVP with search, filtering, sorting, and SEO/Open Graph metadata.

EDUCATION & LANGUAGES
Master's Degree — Automation and Computer-Integrated Technologies | Uzhhorod National University | 2013 –
2019
English — Upper-Intermediate | Ukrainian — Native | Russian — Native