Fractional CTO · SBIR-funded
Fractional CTO for SBIR-funded startups
Turn a Phase I prototype into something Phase II reviewers and customers both believe.
The case
SBIR-funded engineering is not generic engineering.
Most engineering partners have never read a technical volume, let alone written one. I served as principal investigator on NSF SBIR Phase I award #1941226 at Preventa Medical Corporation — $249,251, awarded December 2019 and executed through April 2021 — where I wrote the work plan, ran the project, and filed the project outcomes report. My lane is software: architecture, data pipelines, ML, and the systems around an instrument rather than the instrument itself. What I bring to a grant-funded build is the part most vendors get wrong. A Phase I deliverable is not a demo. It is evidence for a specific technical risk you said you would retire, and it has to be legible to a reviewer who will never run your code.
SBIR founders are usually scientists or clinicians with a real technical insight and no engineering organisation. A full-time CTO is not fundable on a Phase I budget, and a generic dev shop will build what you asked for rather than what the award committed you to demonstrate. A fractional CTO who has held the PI role reads the solicitation the way a reviewer does, and scopes the build to the milestones you are actually accountable for.
The first 90 days
What a sbir-funded engagement actually looks like
A fractional CTO for SBIR-funded work scopes the build to the milestones the award actually commits you to, and architects the Phase I prototype so Phase II extends it rather than replaces it.
- 01
Weeks 1–2: Read the award as a contract
Map every technical objective and milestone in the funded proposal to concrete, buildable work. This is where mismatches surface — objectives that read well in a proposal and describe no testable artifact. Better to find them now than at the annual report.
- 02
Weeks 3–4: Retire the riskiest claim first
Identify the single technical risk the award exists to retire and build the smallest thing that settles it. Phase I fails on unresolved core risk far more often than on unfinished features.
- 03
Weeks 5–8: Build for the Phase II transition
Architect the prototype so the Phase II work extends it. The most expensive outcome in this programme is a Phase I deliverable that has to be thrown away, because it converts a funded head start into a standing restart.
- 04
Weeks 9–12: Evidence and reporting
Instrument the system so results are reproducible and the project outcomes report writes itself from real artifacts. Reviewers weigh evidence, and evidence assembled after the fact is always thinner than evidence captured as you go.
What we cover
SBIR-funded-specific decisions I help you make
Tools I use in sbir-funded
Not ready to talk yet
The Phase I → Phase II readiness worksheet
The questions a Phase II reviewer effectively asks of a Phase I software deliverable, with what a weak answer and a strong one look like. Free, ungated, and most useful around month six — while the answers can still change.
Read it →Request a triage
Talk through your sbir-funded problem.
Free, 30 minutes. Tell me where you're stuck — I'll tell you what it takes. I confirm every request within 24 hours.
FAQ
SBIR-funded questions founders ask
Have you actually held an SBIR award, or just worked on one? +
Held one. Eric was principal investigator on NSF SBIR Phase I #1941226, awarded to Preventa Medical Corporation in December 2019 — $249,251 for a hand-held retinal imager with a flat lens and total internal reflection illumination. The award, the PI name, and the project outcomes report are all public on nsf.gov. That is one Phase I as PI, not a portfolio; the value is that the programme has been experienced from the inside rather than described.
Can SBIR funds pay for outside software development? +
Generally yes — subcontracting and consultant costs are normal, within limits that vary by agency and phase. NSF and most agencies cap the portion of work performed outside the small business, so the split has to be planned rather than discovered. Confirm the current limits in your solicitation and with your grants officer; I scope engagements to fit whatever that constraint turns out to be.
We have not submitted yet. Can you help with the technical volume? +
Yes, and this is the cheapest point to involve an engineer. Most technical volumes are strong on science and vague on the build — reviewers notice, and a work plan written without an engineer tends to commit you to milestones that are awkward to execute. Reviewing it before submission costs a fraction of discovering the mismatch in month two.
Who owns the software? +
You do. SBIR data rights are yours under the programme, and my engagements are work-for-hire on top of that — the client owns the product and the IP. Nothing here creates a dependency you cannot exit.
What happens after Phase I ends? +
Either the prototype becomes the foundation of the Phase II build, or it does not, and that outcome is decided by architecture choices made in the first month rather than at the end. Engagements are structured so the Phase I deliverable is extensible by default; if Phase II is not funded, you still hold a working system rather than a demo.