Solutions Data Scientist, Validations & Backtesting

About the Company

Pilots don't train with real passengers. Actors don't rehearse with real audiences. Yet, the most consequential decisions in society are often pushed straight to production.

Simile is changing that. We have built the first AI simulation of society, populated by generative agents based on real humans. Our research pioneered the field of AI-based simulation, proving it is possible to model human behavior with high accuracy. Today, we are developing a Foundation Model to predict human behavior in any situation, at any scale.

We have raised $300M in total funding, with a $2B post-money valuation in our latest round. Our investors include Greenoaks, Index Ventures, Hanabi, A*, Bain Capital Ventures, Factory, CVS Health Ventures, and Definition.

About the Role

At Simile, accuracy is the product. Enterprise buyers want proof that Simile matches what their customers actually say and do, and as a Solutions Data Scientist on our Deployments team, you will own that proof. You'll design and run customer validations and backtests that compare Simile's simulated results against real-world ground truth, then tune and diagnose until the result is one you can stand behind.

You'll work alongside Solutions Consultants, Deployments, and our research and data science teams, and turn one-off studies into repeatable, scalable workflows. This is the work that most directly builds customer trust in a new category of AI. Your findings will shape where we collect data and how the model improves, on high-stakes validations with some of the world's largest brands.

In this role, you will:

- Study Scoping: Scope validation and backtest studies with Solutions Consultants, Deployments, and customers. Pick the right use case, agree on metrics and success criteria up front, and flag studies that won't produce a meaningful comparison.
- Instrument Replication: Re-field or adapt customer survey instruments so they can be replicated in Simile for a true apples-to-apples comparison, and document every change.
- Simulation & Comparison: Run simulations and compare them against human ground truth using metrics such as TVD, Spearman rank correlation, and top-K / bottom-K agreement for MaxDiff.
- Hill-Climbing with Discipline: Iteratively improve results against a target metric while keeping strict holdout discipline, so a backtest is never overfit to a customer's known answers.
- Diagnosing Misses: Break down gaps by segment and population to determine whether the issue lies in the model, the agent bank, or the ground-truth data itself.
- Reusable Tooling: Build Python pipelines, API-driven experiments, and LLM skills so backtests run reproducibly across clients instead of in one-off spreadsheets.
- Customer Readouts: Write clear readouts showing where the model is strongly aligned and where it is more directional, and present them to technical and business stakeholders.

### Requirements
Must Haves

- Modeling & Hill-Climbing: Experience iteratively improving a model against a target metric — the core of tuning twins until they match real-world outcomes.
- Statistics & Measurement: Comfort with distributional comparison (e.g., TVD), correlation, and choice-based methods like MaxDiff. You know when "accuracy" is meaningful versus noise, and you judge whether a result is real, not just compute it.
- Experimental & Survey Methodology: A strong grasp of sampling, question design, and bias — enough to catch problems in the ground-truth data before blaming the model, including the say-do gap between stated-intent surveys and behavioral data.
- Python & Data Analysis: Fluency running comparisons repeatedly and reproducibly across clients.
- ML Evaluation Discipline: Rigor around holdouts, avoiding leakage, and understanding what pretraining versus post-training changes imply for validity.
- Population & Segment Reasoning: The ability to judge whether a bespoke agent bank actually represents the target audience, and to diagnose misses by segment.
- AI-Native Workflow: Proficiency with Claude Code, Codex, or similar coding agents and working with APIs, which is how we run most experiments, plus experience building LLM skills or automations to scale backtesting.
- Customer-Ready Communication: You can explain methodology, trade-offs, and results clearly to insights, marketing, and data leaders.

Nice to Haves

- Research Background: Experience in market research, consumer insights, survey science, or a research-agency analytics team.
- Survey Platforms: Experience with Qualtrics or similar survey platforms and panel providers.
- Generative Model Evaluation: Experience evaluating LLM or generative model outputs.
- Customer-Facing Technical Roles: Experience in solutions engineering, data science consulting, or analytics consulting.

Compensation & Benefits

At Simile , we provide competitive compensation packages that include base salary, equity, and comprehensive benefits.

- Salary Range: $200,000 – $300,000 USD

Note: Final offers are based on experience, specialized skills, interview performance, and relevant training.

- Equity: Grants are available for eligible roles, subject to board approval.
- Health & Wellness: Comprehensive medical, dental, and vision coverage.
- Time Off: Flexible time off policies to support work-life balance.

Our Process

We prioritize thoughtful conversations and clear examples of past work. Our hiring journey is designed to help both sides align on fit, working style, and expectations.

Reapplication Policy: To ensure a fair and thorough evaluation for all applicants, Simile observes a 90-day waiting period before reconsidering candidates for the same role.

Commitment to Diversity & Inclusion

Equal Opportunity: Simile is an equal opportunity workplace. We welcome applicants of all backgrounds and identities, valuing an environment where everyone can contribute authentically.

Accommodations: If you require support or reasonable accommodations during the application process due to a disability, please let us know. We are happy to assist.