The posting
About monday.com http://monday.com
monday.com http://monday.com is the AI work platform powering the most ambitious teams. 250,000+ customers across departments use us to bring people, workflows, and AI agents together on one flexible platform where AI doesn't just assist, it executes. We move fast, build things that matter, and foster an ownership-driven culture where you're empowered to shape how organizations work and outpace their competition.
About the role
This is a dual-mandate, player-coach role for a senior data science leader who still wants to be close to the hardest technical problems.
Half of your time is hands-on in the Agents group, the team building monday's one of the most strategic AI products. You will be the data science lead for our agents: owning how we define, measure, and improve agent quality, and shaping the technical decisions that determine whether our agents can be trusted to execute real work for hundreds of thousands of customers.
The other half is technical leading the cross-R&D Data Science Guild (~10 data scientists) embedded across product and infrastructure groups. You will set the technical direction, quality bar, and career paths for every data scientist at monday, and make the guild a force multiplier for the whole R&D organization rather than a collection of individual contributors.
Lead data science for the Agents group (hands-on)
- Own the data science agenda for monday's agentic product: what "good" looks like, how it's measured, and how we close the gap between the two.
- Design and build the evaluation stack for agents in production: LLM-as-judge rubrics, trajectory analysis, error taxonomies, regression suites, and post-deployment monitoring, in partnership with AI Infra's centralized Evals framework.
- Drive model, prompt, context, and tool-use decisions with evidence: run the experiments, dig into raw traces, and turn findings into concrete changes the engineering team ships.
- Partner with Agents product and engineering leads on roadmap, tradeoffs (quality vs. cost vs. latency), and launch decisions. Your metrics are the ones the team greenlights on.
- Anticipate the next generation of failure modes as agent architectures evolve, and make sure our evaluation and data strategy stays ahead of them.
Lead the R&D Data Science Guild
- Set the vision and standards for data science at monday: methodology, tooling, evaluation practices, and what "production-grade" means for data scientists embedded in product teams.
- Indirectly manage and develop ~10 data scientists across R&D: performance, growth, career path, and creating a community of practice where knowledge and reusable assets flow between groups.
- Own the hiring bar and pipeline for data science across R&D, and partner with group leads on where the next data science investments should land.
- Act as the senior data science voice in R&D leadership forums, translating what the guild sees into recommendations on product strategy, AI investments, and organizational design.
- Build the mechanisms (reviews, guild syncs, shared platforms, playbooks) that make the guild greater than the sum of its embedded members.
Requirements:
Technical depth in production AI
- 8+ years in data science / applied ML, with significant recent experience building and evaluating LLM-based or agentic systems in production.
- Deep, practical understanding of how agents actually work: models, context management, tool use, harnesses, and where they break. Hands-on with current agent frameworks and SDKs.
- Strong evaluation instincts and experience: you have designed metrics and judges that a team trusted enough to ship on, and you know when a metric is lying.
- Production-grade coding (Python at minimum) and a trace-first diagnostic mindset. You are comfortable in raw logs, notebooks, and pull requests, not only in decks.
Proven leadership of data scientists
- 3+ years managing data scientists, ideally including a distributed or matrixed setup where your people sit inside other teams.
- A track record of raising a team's technical bar: hiring strong people, growing them into senior ICs and leads, and setting standards that stick.
- Comfortable operating as a hands-on player-coach: you protect real hands-on time and model the quality you expect from the guild.
Product and organizational impact
- Strong product intuition. You connect model behavior to customer outcomes and business metrics, and prioritize accordingly.
- Excellent communication with engineers, product leaders, and executives alike, in both technical detail and crisp summary.
- Bias for action in ambiguous, fast-moving environments, and the judgment to know when quality matters more than speed.
- MSc or PhD in Computer Science, Statistics, Data Science, or a related quantitative field.
