NO RED OCEANS

Complaint intelligence for medical device R&D

See how devices fail in the field before you design the next one.

No Red Oceans is building complaint intelligence for medical device research and development teams. We analyze public FDA adverse-event reports from the MAUDE database, along with other lawful public sources, to help engineers and product leaders see how devices fail in use, where unmet needs sit, and which design questions deserve attention early.

Start a conversation How we treat evidence
  • Pre-development
  • Founded 2026
  • Irving, Texas
Illustrative example Written for this page. Not a real MAUDE report.
Source
MAUDE, public
Event type
Malfunction
Reported by
Manufacturer

Event description, as reported

It was reported that the distal tip of the delivery catheter separated1 during withdrawal2. The fragment was retrieved with a snare3. No further patient consequences were reported4.

Structured fields, each tagged by evidence type

  1. 1Device problemComponent separation at the tipEstimate
  2. 2Use contextDuring withdrawalFact
  3. 3InterventionFragment retrieved with a snareFact
  4. 4Patient outcomeNothing further reported; no follow-up in the recordUncertain

InferenceDesign questionDoes the tip joint hold under the loads seen during withdrawal?

What we're building

Three research capabilities, built on public openFDA data. All three are in development. None is a released product, and none has been validated.

  • Search and structure reports

    In development

    Find adverse-event reports by device type, product code, and problem. Turn free-text narratives into structured fields coded to IMDRF adverse event terminology. A machine-made classification stays marked unverified until a person reviews it.

    The question it answersHow does this kind of device fail, and in what situation?

  • Screen for signals

    In development

    Compare how often a problem is reported for one device group against all others, using disproportionality measures such as the proportional reporting ratio, with confidence intervals and minimum case counts. Track how reporting changes over time. A signal is a reason to look closer. It does not show that a device causes harm.

    The question it answersIs this problem reported unusually often for this group of devices?

  • Frame design questions

    In development

    Connect recurring failure modes to the design questions and unmet needs behind them. Each brief lists its sources, date range, method, and limitations, and tags every claim by evidence type.

    The question it answersWhich field problems point to a design study worth doing?

How we treat the evidence

Public complaint data is incomplete and unevenly reported. Our rules start from that.

Fact

Verified fact

What a source actually says, with the source and date. A report stating that a tip separated is a fact about the report, not proof of what happened.

Estimate

Model-derived estimate

A number or classification produced by a model or calculation, shown with its method and assumptions.

Inference

Inference

A conclusion we draw from the evidence, labeled as ours and open to challenge.

Uncertain

Unresolved uncertainty

What the data cannot tell us. We say so instead of filling the gap.

Rules we hold to

  • Complaint counts are not incidence rates. Without a defensible denominator, we describe how often a problem is reported, not how often it happens.
  • No reports does not mean no problem. An absence of reports is not evidence of safety.
  • Complaint volume is not market share. We do not infer one from the other.
  • Every finding shows its work. Source, date range, method, and limitations travel with it.
  • AI output starts unverified. It stays that way until it passes human review.
  • Research and decision support only. We do not give medical, legal, or regulatory advice, and our work does not prove a device safe, unsafe, effective, or defective.

Where we are

We're early, and we'd rather say so plainly. No Red Oceans, LLC was formed in Texas in June 2026. We are writing our quality management system and planning our first product, using research prototypes built on public FDA data. We have no released product.

  1. Stage 1Quality system foundationNow
  2. Stage 2Pre-development planning
  3. Stage 3First product (MVP)
  4. Stage 4Pilot
  5. Stage 5Commercial scale

FounderAlexander Bogdan is a biomedical engineer who spent nearly three years in medical device quality remediation, working with complaint, nonconformance, and CAPA data.

Talk with us

If you design, build, or manage medical devices and have a question that public complaint data might help answer, we'd like to hear it.

Email
LinkedIn page

Please don't email patient information or confidential complaint files.