lvl100Sign in

solstice

Member of Technical Staff (Applied AI, Data Systems)

  • New York City
  • Full-time
  • $200K - $300K
  • Posted

Job description

ABOUT SOLSTICE

Solstice builds AI software that helps pharmaceutical marketing teams turn complex scientific evidence into clear, useful content for healthcare professionals and patients. Our goal is to shorten the time it takes to create and revise that content while making it easy for medical, legal, and regulatory reviewers to trace each claim to its approved source.

We’ve worked with more than 50 pharmaceutical brands and raised a $21 million Series A led by Transformation Capital. Our backers also include Twelve Below, Virtue, and the founders of Datavant, Commure, and Paradigm.

ABOUT THE ROLE

At Solstice, we want the next draft to benefit from what a team learned getting the last one right. We’re hiring a senior engineer to shape our data strategy and build the AI systems that put it to work for pharmaceutical marketing teams. Your decisions will influence what our agents know about a brand, which evidence they draw on, and how expert review improves the content they generate.

Much of that direction is yours to define. Working with product, engineering, and customer teams, you’ll decide which upstream sources are worth connecting first, what to capture from review, and when an investment in shared infrastructure will help multiple teams. You’ll carry those decisions through to production and use what you learn there to set the next priority.

Initially, you’ll spend roughly 60–70% of your time on AI engineering and 30–40% on data engineering. You’ll own the agent workflows and the pipelines behind them from day one. As we connect more upstream sources and the infrastructure becomes more complex, data engineering will take a larger share of your time.

WHAT YOU’LL WORK ON

•Set the direction for data at Solstice. Build a roadmap around the decisions our teams and agents need to make. Establish how teams judge a source’s quality, integrate it, and decide whether it improved the product enough to justify the investment.
•Build on Solstice’s agent platform. Own workflows that turn a brief into a draft, check the draft against its sources, and revise it with a reviewer. Partner with Fullstack AI engineers on shared runtime changes when those workflows need better tool access or tracing.
•Build agent tools for retrieval. Give agents access to current approved content and relevant review decisions. Return the source and version with each result, and enforce the user’s access permissions.
•Connect review decisions to the next draft. Work with product and customer teams to capture what a reviewer changed and why. Use that feedback to improve what the agent retrieves or how it drafts, with changes scoped to the brands and content types where they apply.
•Measure what changes for the people using the product. Evaluate whether drafts use approved evidence accurately and meet the brief. Track whether reviewers spend less time making the same corrections, and use those findings to guide the next engineering decision.
•Build and operate upstream data pipelines. Use Python and SQL to bring briefs, approved materials, and review records from different systems into a common data model. Track which versions were available when a draft was generated so the team can reproduce a decision.
•Keep the data dependable as the platform grows. Detect stale or missing records, handle schema changes, and safely reprocess updates. Own monitoring, access controls, and operating costs as you add integrations.

WHAT YOU’LL BRING

•5+ years of software engineering experience. You’ve owned production systems from design through launch and supported them as people came to rely on them.
•Hands-on AI engineering. You’ve shipped a feature powered by a language model. You can design a tool the model can call, decide what context it needs, and explain which operations belong in application code.
•Experience evaluating AI workflows. You’ve used real tasks to test whether a feature gives users a useful, supported answer. You can follow a trace from retrieval through a tool call to the final response, identify where a failure started, and verify the fix against other cases.
•Production Python and API development. You’ve built and operated Python services that call models, tools, or external APIs. You can handle timeouts and retries, respect access permissions, and use logs and traces to diagnose a failed request.
•SQL and warehouse modeling. You’ve combined data from multiple systems in a cloud warehouse or lakehouse. You can define what each row represents, reconcile identifiers across sources, and preserve version history. You use query plans to find and tune expensive transformations.
•Pipeline design and orchestration. You’ve built and operated incremental ETL or ELT pipelines in Python and SQL. You can manage dependencies between jobs, make writes safe to retry, and backfill late or corrected records without duplicating data.
•Data quality and lineage. You’ve built automated checks for freshness, completeness, and schema changes. You can trace a downstream result to the source versions and transformations that produced it, then identify what needs to be rebuilt when an input changes.
•Judgment and ownership under ambiguity. You’ve taken an open-ended problem from the first conversation to a product people use. You can listen to domain experts, choose a first milestone with incomplete information, explain what can wait, and change direction as you learn.

Helpful experience includes Databricks or Snowflake for data storage and processing, dbt for SQL transformations and tests, and Temporal for pipeline orchestration.

COMPENSATION AND BENEFITS

This is a full-time, on-site role in New York City. Base salary is $200,000 to $300,000, depending on experience and scope of ownership, plus equity. Benefits include health, dental, and vision insurance, unlimited PTO, and a 401(k) match. Visa sponsorship is available for O-1, H-1B, and TN visas.

Browse all jobs by role, city or company