What Hannune Builds

01 · Product

Waitlist · Q3 2026

Entity Resolution API

Entity Resolution as a service. Deduplication and entity linking across datasets — built on Splink with an LLM disambiguation layer. No self-hosting required. Subscribe from $19/mo.

Visit app.hannune.ai →

02 · Platform

Live now

2asy.ai

Graph RAG-powered analytics platform that ingests global economic and geopolitical news daily, extracts causal relationships, and publishes structured analyses with interactive knowledge graphs.

Visit 2asy.ai →

03 · Service

Available now

Consulting

Custom Graph RAG implementation for enterprise teams. Fixed scope, fixed price, fixed timeline. Built on 8+ years of production AI experience across Netflix, Sony, and LG Energy Solution.

See packages ↓

03 · Service Available now

Graph RAG Consulting

Productized Graph RAG engagements with fixed scope, fixed price, and fixed timeline. Built on 8 years of production AI experience across Netflix, Sony, and LG Energy Solution, backed by 2 AI optimization patents. Lean studio model — no agency overhead, direct collaboration with the team building your system.

Package 01 · Validate

Starter

Prove Graph RAG fits your domain before committing to production.

$8,999 fixed

3 weeks · Fixed scope

ScopeSingle data source Graph RAG prototype
StackLangGraph · Neo4j · vLLM · FastAPI
OutputWorking prototype + evaluation report
Incl.Architecture recommendation doc

Book discovery call

Package 02 · Build

Standard

Full production Graph RAG system, deployed and running.

$38,999 fixed

6 weeks · Fixed scope

ScopeMulti-source Graph RAG, production deploy
StackAbove + ElasticSearch · LiteLLM · Docker
OutputProduction system + 1 month hypercare
Incl.Full codebase transfer + docs + training
SLA6-week delivery guarantee

Book discovery call

Package 03 · Scale

Enterprise

Multi-agent systems on your infrastructure, fully owned.

Contact for pricing

Typical range $100K–$250K+

12+ weeks · Custom scope

ScopeMulti-agent architecture + on-prem deploy
StackAbove + Crawl4AI · custom MCP servers
OutputFull system + 3 months on-call support
Incl.Air-gapped option · custom LLM fine-tune
SLADedicated team & support

Request proposal

Add-on · Ongoing

Ops Retainer

Keep the system tuned, updated, and monitored.

Available after any completed package. Includes: monthly optimization pass, ingestion pipeline maintenance, model updates, priority Slack support, and up to 8 hours of additive development per month.

From

$3,499 /mo

Min. 3 months

Custom engagements

Retainer-only arrangements, training programs, or scope beyond the three packages. We reply within 2 business days.

Request a call

Recent Work

Case 01 · Multi-agent pipeline

LG Energy Solution

Enterprise · 2026 Q1

Built a LangGraph-based multi-agent pipeline for talent data acquisition across global sources. Integrated Crawl4AI for structured extraction, EXAONE for domain-specific reasoning, and a custom evaluation framework for output quality.

LangGraph Crawl4AI EXAONE Multi-agent

Case 02 · Domain-specific search

Gahyun Tax Accounting

SMB · Korean tax domain

Developed a Korean-language tax search system for a professional accounting firm. Combined Graph RAG over tax code, precedents, and internal case files with a domain-tuned retrieval layer for fast practitioner lookup.

Graph RAG Neo4j Korean NLP Domain search

2asy.ai — Graph RAG in Production

An automated analysis platform that uses the same Graph RAG pipeline built for clients. Every day, it ingests global economic and geopolitical news, extracts causal relationships, and publishes structured analyses with interactive knowledge graphs.

Currently tracking oil, energy, and geopolitical developments — the pipeline is domain-agnostic and extends to any document-heavy vertical.

2asy.ai preview showing knowledge graph and analysis

~2,400 articles ingested daily
~100 articles selected for knowledge graph
Causal graph visualization on every edition
Runs on Neo4j + LangGraph + Ghost CMS

Visit 2asy.ai →

Reports — Same Pipeline, Different Domains

While 2asy focuses on global macro and geopolitics, the same Graph RAG pipeline works across any document-heavy domain. These are sample analyses on corporate filings, regulatory decisions, and trade policy.

Corporate 60s

NVIDIA 10-K FY2026

Knowledge graph extraction from NVIDIA's SEC 10-K filing using iXBRL parsing.

View →

M&A 60s

Microsoft–Activision: CMA Final Report

How the UK CMA blocked Microsoft's $68.7B Activision acquisition, mapped from the 418-page final report.

View →

Trade Policy 60s

USTR Section 301: 2024 Four-Year Review

US tariff policy on Chinese imports analyzed from the USTR's 187-page review, including 2024 proposed increases.

View →

Live Demos

↗

Three working AI systems, live and interactive. Each demo reflects a real use case (not a prototype).

Ask Hannune ↗

Multi-turn AI assistant that answers questions about Hannune based on a structured profile.

Multi-turn Profile-grounded Conversational

Document Q&A ↗

Upload a PDF and have a multi-turn conversation about its contents using RAG.

PDF Upload RAG Multi-turn

Document Extraction ↗

Upload a PDF and automatically extract structured fields as JSON.

PDF Upload Structured Output JSON

Selected Work

Production systems delivered for enterprise clients.

LG Energy Solution — Talent Data Automation

Replaced manual candidate data collection by building an AI-driven system that navigates complex websites, extracts structured data, and delivers consistent outputs for internal operations. Tech: Crawl4AI, LangGraph, FastAPI, Python.

AI-powered Legal & Tax Document Search

Reduced time-intensive legal document search for a tax professional client. Search time reduced from up to 5 hours to under 40 minutes, with accuracy improved by over 50%. Tech: LangChain, Ollama, ChromaDB, FastAPI, Docker.

Scenera (TnM AI) — Enterprise AI Systems

Delivered multiple AI systems for enterprise clients across data access, IoT pipelines, video analytics, predictive maintenance, and edge deployment. Tech: Python, FastAPI, Docker, LangChain, SQL, Azure.

What Hannune Builds

Hannune builds systems that reduce manual work across data-heavy, search-heavy, and operational workflows.

Workflow automation

Manual processes replaced by automated AI pipelines Large-scale data collection, extraction, and structuring End-to-end document and content workflows

Search and agent systems

Query understanding and retrieval pipelines Extraction, validation, and structured output generation Systems that reduce repetitive document search and review

Production deployment

Docker-based infrastructure for reliable operation REST API integration into client environments Systems running continuously with minimal manual intervention

Identity

Hannune works with enterprise and professional clients who need one accountable technical owner from system design to deployment.

Build and deliver AI systems end-to-end
Own architecture, implementation, and deployment directly
Contract-based work for enterprise and professional clients
Solo by design — centralized responsibility and consistent technical judgment

Operating Model

One technical owner from system design to deployment, with no handoff gaps between planning, implementation, and operation.

Fit

Hannune works best where AI can replace real manual workflows and be deployed into actual operations.

When this is a good fit

Teams spending too much time on manual data collection, document search, review, extraction, or repetitive operational workflows. Enterprise or professional clients who need a production-ready system and want one accountable owner from design to deployment.

When this may not be a fit

Projects centered only on research, experimentation, or vague exploration without a clear workflow, operational problem, or deployment goal may be better served elsewhere.

Get In Touch

If your team is spending too much time on manual data, search, or document workflows, Hannune can design and deliver a production-ready AI system.