Presented to the Board or a committee
Seravune Therapeutics
Board
March 2020 — present
Present research-data context, source limitations and programme status as CDO; executive presentation experience is not statutory director service.
Grace Halim · Chief Data Officer · Singapore / Southeast Asia
I would ask which expert judgement the data presentation is still waiting for.
My contribution to a board
I could help a board understand what a research-data programme has demonstrated and which questions remain for scientific or other specialist judgement. The contribution would concern the context attached to observations, not an independent claim to clinical authority. A common data structure can make values easier to access while concealing different measurement conditions. An archive can replay a stored process without validating its interpretation. I would ask management to explain those boundaries in terms the board can use.
The 2023 cross-study case offers one example. Two measurement methods sat behind 240000 readings from twelve studies. I supported keeping the method families visible pending scientific and statistical comparability review. A board should know whether a combined presentation depends on a conclusion the relevant experts have actually assessed. I would not decide that the methods were equivalent or permanently incomparable from the CDO role alone. The value of the data account is that it gives the responsible judgement a clear basis and identifies where that judgement is still required.
I would also bring the 2011 replicate-reading correction. I had called 2700 readings independent samples when they came from 900 source samples. A board needs to know whether management figures preserve the observation unit or acquire a larger meaning as they move into a presentation. My own mistake would make that challenge less abstract. The correction retained the useful readings and changed the unsupported interpretation, with statisticians responsible for the actual inference.
This proposed contribution could fit a research organisation improving its evidence practices, but it would need to be assessed alongside the board's existing expertise. I claim no immediate readiness to chair a specialist committee and no statutory service in this narrative. A first appointment would require learning, permission and a realistic understanding of the collective role, including where another director or adviser should lead the discussion.
Board and committee exposure
Committee contribution
| Capacity | Committees |
|---|---|
| Ready to join | Data or research oversight committee considering the provenance and usability of institutional evidence |
Capacity and preferences
- Time available
- 24 days per year
- Current directorships
- 0
- Sectors
- Therapeutics research organisations · Research-service businesses · Established healthcare organisations improving their research-data practices
- Company type
- An established private research group with clearly separated scientific, clinical and data responsibilities
- Company stage
- An institution improving data discovery, cross-study interpretation or analysis archives
The seat I am pursuing
I would consider a first non-executive opportunity where the board needs a practical understanding of research-data context and the limits of programme evidence. The institution should value clear specialist responsibility and an honest account of unresolved work. Twenty-four annual days are conditional on employer permission and the actual demands of preparation, induction, committee work and travel. I would seek a contribution matched to the board's needs and my further learning, rather than presume that a CDO title provides clinical expertise or readiness to chair a scientific committee. Any appointment would need its own current professional assessment.
Judgement in practice
The missing field stayed missing until its meaning could be examined
The 2021 registry had 11200 recorded baseline values and 6800 missing entries within 18000 records. I opposed coding the missing portion as normal and supported source and fitness review. I would ask a board whether preparation has added an interpretation the source does not support. Missingness can have different origins, and the responsible experts need to determine how that affects the intended analysis. The decision made the table less convenient but kept the uncertainty visible. It did not produce a clinical conclusion about the missing records or establish that every absent value was later recovered.
A practical retrieval result stayed within its measured scope
The 2017 metadata work covered thirty-seven datasets and a six-month comparison of twenty-six matched request pairs. Median retrieval changed from eleven to four business days. I would present that as a bounded internal service result, with research, data and technology teams contributing. It does not show that metadata alone caused the change or that a scientific or patient outcome improved. The board question is whether the programme's actual evidence supports the benefit management is claiming. A useful result should be recognised without extending it beyond the comparison that produced it.
Replay status did not become scientific validation
Forty-nine of sixty-four archived analyses replayed in the 2025 exercise, while fifteen lacked a required dependency version. I retained that split and supported S$2.9 million remedial work. A board could ask what successful replay establishes, how the unresolved cases are owned and which scientific questions remain outside the exercise. The code running again does not independently validate every analytical choice or conclusion. I would help make that distinction understandable so that a technical success receives appropriate credit without being allowed to settle a broader question it did not test.
My first year on a board
I would begin by understanding how the board receives evidence about research-data programmes and which specialists provide the relevant assurance. I would read the committee remit, learn the institution's collaboration and ownership context, and ask how measurement methods, missing source material and analytical changes are made visible to management. The aim would be to understand the existing arrangement before proposing a new reporting mechanism or data initiative.
Working with fellow directors, I would identify a small number of questions where my experience could add practical scrutiny. A combined portfolio view or an archive programme might be suitable starting points. I would seek the responsible scientific, statistical, clinical or other specialist view and avoid creating a parallel executive data function. I would also review my own learning and preparation, including whether the proposed time remains adequate. The contribution should help the board distinguish process progress from the conclusions that depend on it, while retaining clear accountability for both.
Conflicts and their management
Research collaborations, employment, suppliers and investor relationships would need confidential disclosure through the proposed institution's process. A public profile cannot substitute for that review or safely publish every relevant relationship detail. I would ask what permissions and continuing assessments are required and be prepared to decline a role if appropriate separation is not possible. The fact that a contribution concerns data rather than a therapeutic decision does not remove the need to examine those relationships. This fictional account establishes no finding of independence and no permission to share restricted research or participant material with another institution.
My independence statement
Employment, research collaborations, suppliers and investor relationships would need disclosure and review for the specific institution. This fictional account cannot establish independence, and presenting research-data matters to a board does not settle the conditions of a director appointment.
Governance learning and readiness
IICA Independent Directors’ Databank: Not yet registered
The fictional Singapore-based career records no Indian independent-director databank registration and makes no exemption claim. Any actual appointment would require current guidance on the relevant jurisdiction, institution and responsibilities.
- Develop understanding of collective director responsibility beyond the executive task of preparing a research-data account.
- Study the proposed institution’s financial reporting, scientific assurance and the separation of clinical, regulatory and data authority.
- Learn how its board evaluates research evidence, unresolved source limitations and the responsibilities of collaborating organisations.