Evolrix AI — Making biology computable — AI for synthetic biology manufacturing

Method

One loop. Every step labeled.

Kairos runs the design–build–test–learn cycle as a dry-lab loop today. Each stage declares what it is — retrieved evidence, model reasoning, or a deterministic tool. Nothing in between.

01AskNatural-language research questionLive

02RetrieveCited evidence, with provenanceLive

03DesignDeterministic CDS & leader panelsLive

04RankHost-fidelity scores you can re-runLive

05FermentWet-lab validation flows backNext

Solid = running today · Dashed = planned, shown as planned

Product

Watch one Kairos session.

Every number below comes from the published cytochrome-c worked example — nothing staged.

kairos · live session

See the full worked example

Evidence

Numbers we publish — and re-run.

Retrospective dry-lab metrics for Kairos v0.4.0, reproducible from the deterministic tools — including the uncomfortable ones.

ρ ≈ +0.35

Within-protein host-fidelity ranking (Spearman, LOGO cross-validation)

ρ ≈ 0.12–0.26

Cross-source generalization, method-dependent — reported, not rounded up

627 / 21

Expression-construct records / cross-source folds behind the ranking model

0

Expression or titer predictions made. Real expression awaits wet-lab data

Methods & benchmarks

Honest scope

What Kairos does. What it does not claim.

Kairos does

Dry-lab design quality — ranked and cited.

✓Ranks synthesis safety & host resemblance for Pichia constructs
✓Optimizes expression-construct elements — learned effects matching known biology (pCS1, ERO1/SBH1)
✓Scores 5′ translation initiation & protein developability
✓Checks structures in silico — AlphaFold DB + Boltz-2 confidence
✓Cites every claim to literature you can open and check

Kairos does not

On the record, so you can hold us to it.

—No expression or titer prediction — wet-lab outcomes await wet-lab data
—No host-agnostic claims — Pichia-scoped until other hosts validate
—No opaque scoring — every score re-runnable over the printed sequence
—No invented evidence — where data is thin, the dossier says so

Questions

What teams ask.

What is Evolrix AI?

An AI company building the tools that make synthetic biology computable — dramatically accelerating the path from research question to manufacturing-ready design. Designs that are provably correct, scores you can re-run, and answers grounded in literature. Built for teams that need provenance, not persuasion.

Who is it for?

Industrial R&D teams in synthetic biology, metabolic engineering, and biotech — from early-stage design to pre-manufacturing validation.

Is Kairos an expression or titer predictor?

No. Kairos ranks dry-lab design quality — synthesis safety, host resemblance, and expression-construct optimization (learned element effects matching known biology (pCS1, ERO1/SBH1); cross-source generalization ρ≈0.12–0.26 (method-dependent)), 5′ translation initiation, and protein developability. It is not an expression or titer prediction. Real expression awaits wet-lab data.

What can it do today?

Kairos v0.4 (Pichia-scoped): conversational research with cited evidence retrieval, LLM-backed reasoning (Qwen3), ranked CDS design panels with host-fidelity scores and expression-construct optimization (element effects matching known biology (pCS1, ERO1/SBH1); ~627 records, 21 folds), 3D protein structure prediction via AlphaFold DB, and secretion leader design. Expanding to more hosts.

How does the chat work?

Ask a research question in natural language. Kairos retrieves cited evidence, reasons with an LLM, runs deterministic design tools, and returns a structured dossier — streamed live. Each section carries its provenance: evidence (retrieved, cited), reasoning (model or templated), design (deterministic tools).

Bring one target protein. Leave with a ranked panel.

Minutes, not weeks. Every construct scored, every claim cited, every limit stated.

Try Kairos Request a scoped pilot

Latest research