Autonomous experiments for programmable biology

Design is scaling faster than experimental proof

• Models generate candidates continuously — proteins, sequences, constructs, and experiment plans are now computational outputs
• Physical evidence remains scarce — instrument access, method transfer, scheduling, and manual data handling constrain iteration
• Cloud labs validate the model — Ginkgo and Emerald show that researchers will access laboratory infrastructure remotely
• The next layer is programmable execution — intent becomes lab operations; results return as structured evidence for the next design cycle

The first Hyderabad cell is operational

The technical question is no longer whether a cell can run. The financing question is how quickly it can become reliable customer capacity.

Close the design–execute–learn loop

Selected programs begin through a reviewed expression of interest. Standardized methods move into transparent self-serve ordering and API access as the fleet scales.

LabOS connects intent to instruments

• Protocol translation — approved experimental intent becomes versioned instrument operations
• Capability and quote — requests map to validated methods, materials, capacity, and price
• Scheduling — plates, instruments, and incubation windows become an optimized execution plan
• QC and exceptions — controls, telemetry, interventions, and re-run decisions are explicit
• Experimental graph — designs, operations, measurements, and outcomes remain linked for reuse

Match public prices. Win on execution.

PLANNING ASSUMPTIONS Targets benchmarked to Ginkgo's public catalog on 31 July 2026. They are not historical Programmable Bio prices, availability claims, or binding quotes.

Three revenue layers

• Per-assay pilots — reviewed programs and standardized launch catalog
• Reserved capacity — recurring batches and service levels for industry design loops
• Platform and network — API access, method portability, and future multi-node execution

PLANNING ASSUMPTIONS Replace with measured throughput, labor, consumables, failure, maintenance, and CapEx data before formal fundraising use.

Size the first market from reachable users

PLANNING ASSUMPTIONS Illustrative customer counts and spend; replace with sourced account lists, interviews, EOI conversion, and observed order frequency.

Begin with pilots, standardize what repeats

• Demand signal now — customer EOI captures assay, batch, timing, and budget before further CapEx
• Catalog discipline — only standardized, measured workflows receive public prices and service levels
• Scale by repetition — add instruments and cells where accepted demand and utilization justify them

Direct competition is already strong

Primary sources and verification dates are maintained on the public landscape page.

From working cell to customer fleet

PLANNING ASSUMPTIONS Dates are targets, not commitments. Apollo work begins only after a signed agreement and approved governance.

Apollo is proposed, not contracted

• No signed MoU today — no EHR, biospecimen, patient, exclusivity, or branding rights are claimed
• Strategic rationale — a future collaboration could connect reproducible autonomous assays to well-governed, population-relevant research cohorts
• Required gates — agreement, ethics review, consent, data governance, security, IP, and approved public language
• Business is independent — Hyderabad lab scaling and customer validation do not depend on Apollo closing

An illustrative $3M seed to scale execution

PLANNING ASSUMPTIONS $3M, 24-month runway, 3 expanded cells, 10 qualified assays, and 25 paying customers are draft planning targets. Add named founder bios and replace all figures before formal circulation.