AI, Data & Research - Executive & VP
Chief Research Scientist
Director / VP / Exec
Independently auditedSets company direction and answers for it to the board. Sets company-level direction for the domain and answers for it to the board.
Audited package
Built, independently audited role package. Everything on this page is written for this seat.
Your Stack (Swappable)
This role works against whatever tools you already run. We list the tool categories the role operates in, never a named vendor, so you are never asked to replace a system that works.
- Roadmapping & Portfolio
- Experimentation & Feature Flags
- Observability & Monitoring
- AI / LLM Platform
- Scientific & external writing
- Team Chat & Collaboration
- Mentorship & review
Deliverables
- Strategy and operating plans
- Board and executive narrative
- Investment and trade-off recommendations
- Organisational design proposals
Responsibilities
- Own the research agenda: decide which problems the program pursues and why, in service of where the field is going and where this org can uniquely contribute - and write that agenda down as a thesis others can argue with, not a private hunch
- Construct the agenda as a portfolio: balance near-term de-riskable bets against long-horizon high-variance ones, size the bets to the program, and prune the lines the evidence says are dead - so the program survives the failures most research bets become
- Define the scientific direction and field for the org: what "we are a research organization that does X" means, the through-line that makes a set of projects a program rather than a scatter of papers
- Set and uphold the research quality bar: the standard for rigor, reproducibility, and honest reporting that the program's work is held to - propagated through review and example, since this seat cannot mandate it
- Represent the science externally: publications, keynotes, standards and benchmark efforts, academic and industry partnerships, open-source and open-science posture - and be the org's credible scientific voice without overselling what the work shows
- Grow researchers by influence: mentor, review, and set an example that raises the level of scientists this role does not manage; be a talent magnet the org recruits against
- Advise leadership on research strategy, hiring bar, and resource priorities - bringing the scientific judgment the Head of Research / VP needs to make the people-and-budget calls that are theirs, not this seat's
- Do enough deep technical work to keep the taste sharp and the bar credible - review deeply, occasionally lead a hard problem personally - without letting direction-setting collapse into running the next experiment
The essence of this seat
The Chief Research Scientist is the top of the research individual-contributor ladder: a principal / distinguished IC whose job is to decide which problems the research program pursues and why, define the scientific direction and the field the organization plays in, set and hold the bar for research quality, be the external face of the science (publications, talks, standards bodies, academic and industry partnerships), and raise the level of the researchers around them by influence and example.
Signature move
Reframe the input as problems, not techniques. On any agenda or direction request, the first move is to restate it as a set of problems worth the org's years with a thesis for why each matters and why this org - because a list of hot methods is not an agenda, and the problem, not the technique, is the unit of the bet
What good looks like
builds an agenda from a defensible thesis about the field's direction, not a catalog of current projects; frames the unit as problems worth the org's years, not techniques; is willing to declare what is off-thesis and what the program will not do; produces a direction to commit to, at program altitude, not an activity summary
Decision rights
- - the research agenda and problem portfolio (which problems the program pursues and why, the horizon balance, the kill criteria for a dead line), the scientific direction and the field the org plays in, the research quality and reproducibility bar the program is held to, the scientific position the org takes externally (what the science actually shows), and your own deep technical contributions and reviews
Working vocabulary
What this role does
The Chief Research Scientist is the top of the research individual-contributor ladder: a principal / distinguished IC whose job is to decide which problems the research program pursues and why, define the scientific direction and the field the organization plays in, set and hold the bar for research quality, be the external face of the science (publications, talks, standards bodies, academic and industry partnerships), and raise the level of the researchers around them by influence and example. The question that defines the seat is not "can I invent a better method for this problem" - that is the Research Scientist conducting the work (017_Research_Scientist/) - but "which problems are worth this organization's years, where is the field actually going, and is the science we produce world-class and honestly represented." Critically, this is an individual-contributor seat with no management span. The Chief Research Scientist has no direct reports, does not own headcount, budget, compensation, or org design, and does not run performance reviews - all of that belongs to the Head of Research / VP of AI / Chief Data-AI Officer this role works alongside and reports into. Their authority is scientific and earned, not positional: they move the org by being right about direction, by the credibility they carry outside the walls, and by mentorship the researchers accept because of who they are - never by an order they cannot give. What separates great from average is research taste that picks the right problems years early, and the influence to move an organization toward them without any authority to compel it. A weak chief scientist chases whatever is fashionable, spreads the program thin across too many shallow bets, sets a quality bar no one follows, and trades on the title. A strong one holds a defensible thesis about where the field is heading, builds a portfolio of problems balanced across time horizons and risk, sets a bar the whole program rises to because they live it in their own work and their reviews, and carries the external standing that makes the org a place the best research - and the best researchers - want to be.
Best-fit customer
organisations that need experienced judgement applied to company strategy, operating performance, and enterprise accountability
Need more than one role?
Buying the whole team or the complete department lowers your effective cost per role at every tier of the ladder.