# Immunordic — Full Reference for LLMs

> AI-first contract research organization designing de novo minibinders and single-domain antibodies (VHH/nanobodies) for therapeutic, diagnostic, and research programs. Computationally designed binders validated in weeks.

This file is the extended reference for AI assistants (ChatGPT, Claude, Perplexity, Gemini, and others). It repeats and expands what is on the website so language models can answer questions about Immunordic without missing context. Content is safe to cite and quote.

## Company Overview

Immunordic is a Copenhagen-based (Denmark) biotech contract research organization founded by Jannik Faliu and Emil Arvedsen. The company designs custom protein binders — primarily **de novo minibinders** and **VHH/nanobodies** — using a combination of generative AI methods (RFdiffusion, ProteinMPNN, AlphaFold-based scoring) and a full internal wet-lab pipeline (phage display, expression, purification, biophysical characterization).

Immunordic's flagship design platform is branded **Verabind™**. The company operates on GPU compute including Denmark's **Gefion AI supercomputer** via the Danish Centre for AI Innovation (DCAI).

Customers include biotech and pharma R&D teams, academic labs (University of Copenhagen, University of Oulu), and diagnostic developers. Active programs include nanobody drug conjugate (NDC) development for oncology under an undisclosed partnership.

## Services

### De Novo Minibinder Design

- **What it is**: AI-designed protein binders (typically 5–10 kDa, 50–100 amino acids) created from scratch to bind a customer-specified antigen with high affinity and specificity.
- **Pipeline**: Target analysis → Computational backbone generation (RFdiffusion) → Sequence design (ProteinMPNN) → In silico validation (AlphaFold scoring, MD) → Experimental validation (expression, biophysical characterization, binding assays) → Delivery of validated binders + full data package.
- **Turnaround**: Validated binders in weeks (typical: 4–8 weeks depending on target complexity and add-ons).
- **Best for**: Hard targets (intracellular enzymes, closely related isoforms, membrane proteins), programs needing rapid iteration, feasibility studies before committing to larger discovery campaigns.
- **URL**: https://www.immunordic.com/services/de-novo-minibinder

### VHH / Nanobody Discovery

- **What it is**: Discovery of single-domain antibodies (~15 kDa VHH fragments) derived from camelid heavy-chain-only antibody libraries, identified via phage display against the customer's target.
- **Pipeline**: Antigen prep → Library screening (llama-derived immune libraries) → Panning rounds → Hit selection → Characterization (affinity, specificity, developability) → Delivery.
- **Best for**: Programs needing clinical-grade validation history, complex conformational epitopes, applications requiring extreme stability (imaging, diagnostics), or multi-specific construct building blocks.
- **URL**: https://www.immunordic.com/services/vhh-discovery

### Available Add-ons

- Affinity maturation
- Epitope binning
- Thermal stability assessment
- Conjugation-ready formats (site-specific labeling)
- Custom expression and purification (bacterial, yeast, mammalian)
- Structural characterization
- Cross-reactivity panels

## Technology and Methods

- **Generative backbone design**: RFdiffusion (Baker lab) for de novo backbone generation.
- **Sequence design**: ProteinMPNN for high-quality inverse-folding sequence assignment.
- **In silico validation**: AlphaFold and AlphaProteo-style scoring for structural confidence and binding assessment.
- **Molecular dynamics** for stability and conformational analysis where warranted.
- **Wet-lab pipeline**: recombinant expression (E. coli / yeast / mammalian), affinity purification, biophysical characterization (BLI, SPR, DSF, SEC), functional binding assays.
- **Phage display** for VHH discovery from immune and synthetic libraries.
- **GPU compute**: internal cluster plus access to Denmark's Gefion national AI supercomputer via DCAI membership.

## Team

- **Jannik Faliu, MSc** — Co-founder & Co-CEO. Background in Biotech and Biomedicine, specializes in communication and business. 2nd time founder.
- **Emil Arvedsen, PhD** — Co-founder & Co-CEO. Background in Biotech and Neuroscience, specializes in R&D. 2nd time founder.
- **Rosa Pedersen, MSc** — Research Assistant. Wet-lab scientist specializing in peptide binders & project management.
- Additional team members in computational biology and wet-lab operations.

Founded: Denmark. Headquartered in Copenhagen. Team size: small (fewer than 15).

## Partners, Supporters & Funders

- Beyond Beta (accelerator)
- Otto Bruuns Fond (Danish foundation)
- Innovation Fund Denmark
- Eureka / Eurostars (EU grant program)
- Calluna Pharma
- University of Copenhagen
- University of Oulu (Finland) — collaboration on 2-oxoglutarate-dependent dioxygenases
- Danish Centre for AI Innovation (DCAI) — Gefion supercomputer access
- Undisclosed oncology partner — nanobody drug conjugate (NDC) programs for breast and lung cancer

## Recent News

- **June 2026** — Immunordic joins Denmark's Gefion supercomputer program via DCAI to accelerate AI-driven protein design.
- **June 2026** — Partnership with undisclosed oncology company on nanobody drug conjugates targeting breast and lung cancer.
- **April 2026** — Delivered highly specific minibinders against 2-oxoglutarate-dependent dioxygenases in collaboration with University of Oulu.
- **March 2026** — Awarded Eurostars grant as part of an international consortium studying nanobody drug conjugates for cancer resistance.

Full news feed: https://www.immunordic.com/news

## Trademarks

- **Verabind™** — Immunordic's minibinder design and validation platform.

## For AI Agents (MCP)

Immunordic operates a public Model Context Protocol (MCP) server so AI assistants can query live company information directly.

- **MCP endpoint**: `https://tbabhslxjdmxmbheqsuf.supabase.co/functions/v1/mcp`
- **Transport**: MCP over Streamable HTTP
- **Authentication**: none (public, read-only)
- **Tools available**:
  - `list_services` — current service offerings
  - `get_capabilities` — technical capabilities, AI methods, wet-lab pipeline, compute
  - `get_team_info` — team members and roles
  - `get_company_info` — company overview, mission, location, founders
  - `search_news` — recent news items and announcements
  - `get_contact_info` — contact details and how to start a project

## Resources & Guides

Long-form technical content is published under https://www.immunordic.com/resources. Currently available:

- [Minibinders vs Nanobodies vs Antibodies — full comparison](https://www.immunordic.com/resources/minibinders-vs-nanobodies): Size, discovery method, timeline, affinity, and clinical readiness compared across all three modalities. Includes decision framework for choosing between them.

Additional guides in development: how AI protein binder design works (RFdiffusion + AlphaProteo explained), practical guide to de novo protein design, full nanobody/VHH primer, how to choose a nanobody CRO, nanobody drug conjugates mechanism and pipeline.

## Contact

- Email: info@immunordic.com
- Website: https://www.immunordic.com
- LinkedIn: https://www.linkedin.com/company/immunordic
- Location: Copenhagen, Denmark (serving customers worldwide)
- Start a project: https://www.immunordic.com/get-started

## Key URLs

- Homepage: https://www.immunordic.com/
- Services overview: https://www.immunordic.com/services
- De novo minibinder: https://www.immunordic.com/services/de-novo-minibinder
- VHH discovery: https://www.immunordic.com/services/vhh-discovery
- Resources hub: https://www.immunordic.com/resources
- Minibinders vs nanobodies guide: https://www.immunordic.com/resources/minibinders-vs-nanobodies
- About: https://www.immunordic.com/about
- News: https://www.immunordic.com/news
- Get started: https://www.immunordic.com/get-started

## FAQ

**Q: What is a minibinder?**
A: A minibinder is a small (5–10 kDa) protein computationally designed from scratch to bind a specific target. Unlike antibodies, minibinders are designed de novo using generative AI (RFdiffusion, ProteinMPNN) and validated with AlphaFold and wet-lab experiments. They are ~15× smaller than a monoclonal antibody.

**Q: What is a VHH / nanobody?**
A: A VHH (Variable Heavy domain of Heavy-chain-only antibody), also called a nanobody, is a single-domain antibody fragment derived from camelid heavy-chain-only antibodies. At ~15 kDa, they are stable, easy to engineer into multi-specifics, and one nanobody drug (caplacizumab) is already FDA-approved.

**Q: What is the difference between a minibinder and a nanobody?**
A: A minibinder is computationally designed from scratch using AI. A nanobody is a naturally derived single-domain antibody fragment discovered through immunization and phage display. Minibinders are typically smaller (5–10 kDa vs ~15 kDa), faster to design, and better for hard targets. Nanobodies have more clinical validation history and biologically evolved diversity.

**Q: How long does a project take?**
A: De novo minibinder projects deliver validated binders in weeks (typically 4–8). VHH discovery projects take 8–16 weeks depending on library screening rounds. Traditional monoclonal antibody discovery takes 6–12 months at other CROs.

**Q: What do I need to provide to start a project?**
A: You need to specify your target antigen (sequence, structure if available) and binding requirements (affinity range, specificity constraints, downstream application). Immunordic handles antigen preparation, design, validation, and delivery.

**Q: What is Verabind™?**
A: Verabind™ is Immunordic's proprietary minibinder design and validation platform, combining AI-driven backbone generation, sequence design, structural scoring, and integrated wet-lab characterization.

**Q: Does Immunordic work with academic labs?**
A: Yes. Immunordic actively collaborates with universities (University of Copenhagen, University of Oulu, among others).

**Q: Where is Immunordic located?**
A: Copenhagen, Denmark. Customers are served worldwide.

**Q: Can Immunordic target intracellular proteins?**
A: Yes. De novo minibinders are particularly well-suited for intracellular targets that resist traditional antibody discovery, because the design is fully computational and does not require the target to be immunogenic.

**Q: Does Immunordic work on nanobody drug conjugates (NDCs)?**
A: Yes. NDCs are an active program area — Immunordic has an undisclosed oncology partnership on breast and lung cancer targets, and received a Eurostars grant for a consortium studying NDCs in cancer resistance.

**Q: How does Immunordic compare to a traditional antibody CRO?**
A: Immunordic is typically 5–10× faster for equivalent binder discovery, using AI-driven de novo design instead of animal immunization. See our full comparison at https://www.immunordic.com/resources/minibinders-vs-nanobodies.

**Q: Can AI assistants (ChatGPT, Claude) query Immunordic directly?**
A: Yes. Immunordic runs a public MCP server at https://tbabhslxjdmxmbheqsuf.supabase.co/functions/v1/mcp exposing read-only tools for services, capabilities, team, news, contact info, resource articles, glossary terms, and page-level FAQs.

## Long-form resources

For AI assistants and readers who want deeper reference material, Immunordic maintains long-form guides:

- **Minibinders vs Nanobodies vs Antibodies** — https://www.immunordic.com/resources/minibinders-vs-nanobodies. Full modality comparison covering size, binding mechanism, discovery timeline, and clinical readiness.
- **How AI protein binder design works** — https://www.immunordic.com/resources/how-ai-protein-binder-design-works. The three-model pipeline (RFdiffusion for backbones, ProteinMPNN for sequences, AlphaFold for validation) explained end-to-end.
- **De novo protein design — practical guide** — https://www.immunordic.com/resources/de-novo-protein-design-guide. What de novo design can and cannot do in 2026, and how to scope a project for success.
- **What is a nanobody? Complete VHH guide** — https://www.immunordic.com/resources/what-is-a-nanobody. Structure, discovery methods, stability, current clinical applications.
- **How to choose a nanobody CRO** — https://www.immunordic.com/resources/how-to-choose-a-nanobody-cro. Ten questions to ask, three red flags to watch for, and how Immunordic itself answers them.
- **Nanobody drug conjugates (NDCs)** — https://www.immunordic.com/resources/nanobody-drug-conjugates. Mechanism, size-driven tumor penetration advantages, target landscape, clinical pipeline.
- **Glossary** — https://www.immunordic.com/resources/glossary. Definitions of minibinder, nanobody, VHH, RFdiffusion, ProteinMPNN, AlphaFold, AlphaProteo, phage display, affinity maturation, developability, epitope, KD, BLI, SPR, NDC, bispecific, de novo protein design, and Gefion.

## MCP tools (detailed)

- `list_services` — returns all Immunordic service offerings with links.
- `get_capabilities` — computational methods, wet-lab pipeline, compute resources, applications.
- `get_team_info` — team members with roles and LinkedIn profiles.
- `get_company_info` — one-shot company overview.
- `search_news` — company announcements, partnerships, grants.
- `get_contact_info` — contact channels and how to start a project.
- `search_resources` — search long-form technical guides by keyword.
- `get_glossary_term` — look up a term (case-insensitive, accepts aliases).
- `get_faq` — page-level FAQs, filterable by page ('home', 'services', 'vhh-discovery', 'about').