HEALTHCARE
Dermsquared Polaris
Real-time clinical answers for dermatology professionals
Learn how AE Studio built Polaris, an AI-augmented search engine trained on 20+ years of Dermsquared proprietary content, delivering real-time clinical answers and intelligent content recommendations to dermatology professionals.
AE Studio is technology- and innovation-driven.
Lucia Stetson, Chief Transformation Officer, Dermsquared
THE CHALLENGE
The problem.
Dermatologists face a daily problem of information overload. Staying current with treatment protocols, drug interactions, access management workflows, and emerging research requires constant effort. When a clinical question arises mid-appointment or during documentation, there is no fast, reliable way to search a trusted dermatology knowledge base without context-switching to general search engines that surface generic, unvetted results.
Dermsquared had spent over 20 years building one of the most comprehensive repositories of dermatology-specific content, including expert videos, podcasts, conference proceedings, CME materials, case studies, and drug documentation. But that content was fragmented across the platform and not optimized for fast, conversational retrieval. Users could browse or scroll, but not simply ask a clinical question and get a direct, authoritative answer.
Traditional keyword search returns lists of results, but dermatologists need answers. A physician asking about systemic medication approval workflows or appropriate treatments for a teenage acne patient needs a synthesized response grounded in vetted clinical content, not ten links to sort through manually.
Dermsquared needed a search experience that could function as a digital clinical assistant, one that understood the intent behind clinical queries, pulled from verified expert content, and surfaced related resources in a way that felt native to how physicians actually work.
THE SOLUTION
What we built.
AI-Augmented Search Built on Proprietary Clinical Content
Standard search engines are not built for clinical precision. AE Studio built Polaris as an AI-augmented search engine specifically trained on Dermsquared's proprietary content library, accumulated over more than two decades of expert-driven dermatology education.
The system goes beyond keyword matching. Polaris interprets the intent behind clinical questions, categorizes the query type in milliseconds, and routes it to the appropriate knowledge base before generating a synthesized, clinically grounded response.
Query Classification and Specialized Routing
Not all clinical questions are the same. A question about Chronic Spontaneous Urticaria symptoms requires a different retrieval strategy than a question about navigating prior authorization for systemic medications.
Polaris uses precision-targeted algorithms to classify queries into categories, including clinical questions, practice management topics, access management workflows, and conference-specific content. Each category triggers a specialized prompt strategy, ensuring that the response is calibrated to the nuance of what the physician is actually asking.
This classification layer also enables Polaris to recognize when a question falls outside its knowledge base and return an honest, non-hallucinatory response rather than fabricating an answer.
Hybrid AI and Keyword Search
Polaris combines AI-generated responses with traditional keyword search results, giving physicians both a synthesized answer and access to all related source material. This hybrid approach preserves the depth of Dermsquared's content library while providing the immediacy that clinical contexts demand.
Search results surface related articles, videos, podcasts, and conference materials alongside the AI response, creating a seamless experience where users can go deeper on any topic without leaving the platform.
Automated Content Ingestion Pipeline
Clinical knowledge evolves constantly. A search engine trained on static content quickly becomes outdated. AE Studio built an automated content pipeline that continuously indexes new dermatological content as it is published on Dermsquared, including news articles, research findings, podcast transcripts, video content, CME materials, case studies, and drug documentation.
This pipeline ensures that Polaris always reflects the current state of Dermsquared's knowledge base without manual intervention. When new expert content is added, it becomes searchable through Polaris automatically.
Clinical Accuracy Guardrails
In clinical settings, inaccurate information is not just unhelpful, it can be harmful. AE Studio implemented multi-layer verification within Polaris to validate AI responses against trusted dermatological sources and eliminate hallucinations.
The system is designed to recognize the boundaries of its knowledge. Polaris only answers questions it can ground in verified Dermsquared content. Non-clinical or out-of-scope queries are handled gracefully, with the system indicating when a topic falls outside its knowledge base.
Polaris v2: Expanding to the Full Knowledge Base
Following the successful v1 launch, AE Studio partnered with Dermsquared to expand Polaris to cover 100% of the platform's public-facing content. Polaris v2 extended content mapping beyond the initial library of articles and media to include the entire Dermsquared website, from practice management pages to condition-specific resources.
V2 also introduced a restructured prompt strategy that identifies the category of a search query first, then applies a specialized prompt for that category. This architecture allows future capabilities to be layered in without disrupting established features, making Polaris extensible as Dermsquared's content and physician needs continue to evolve.
Content Diversity and Related Resource Discovery
Beyond delivering direct answers, Polaris functions as a content discovery layer across the Dermsquared platform. After returning a clinical response, the system surfaces related videos, podcasts, research articles, and conference content, keeping physicians connected to the broader Dermsquared knowledge ecosystem.
This related content surfacing increases dwell time on the platform, drives engagement with content types physicians might not have discovered through browsing, and reinforces Dermsquared's value as the definitive professional resource for dermatology.
HOW IT WORKS
The details.
A Search Engine Trained on 20 Years of Dermatology Content
Standard search engines are not built for clinical precision. AE Studio built Polaris specifically on Dermsquared's proprietary content library, built up over more than two decades of expert-driven dermatology education. Polaris does not just match keywords. It interprets the meaning behind a clinical question and routes it to the right knowledge before generating a response.
Different Questions Get Different Treatment
A question about a skin condition requires a different approach than a question about navigating prior authorisation for a medication. Polaris classifies each query in milliseconds and applies a tailored strategy for that category. This means each answer is calibrated to the specific type of question being asked, not just the surface-level words. When a question falls outside the knowledge base, Polaris says so rather than making something up.
AI Answers Alongside Traditional Search Results
Polaris combines a synthesised answer with traditional search results. Physicians get both a direct response and access to all related source material: articles, videos, podcasts, and conference content. They can go deeper on any topic without leaving the platform.
New Content Becomes Searchable Automatically
Clinical knowledge changes constantly. A search engine trained on static content goes out of date. We built a pipeline that continuously indexes new content as it is published on Dermsquared. When new expert material is added, it becomes searchable through Polaris without any manual steps.
Guardrails That Prevent Harmful Answers
In a clinical setting, inaccurate information is not just unhelpful. It can cause harm. Polaris validates every AI response against trusted dermatological sources. If the answer cannot be grounded in verified Dermsquared content, Polaris indicates that rather than inventing an answer. Non-clinical questions are handled gracefully with a clear indication that they fall outside the scope of the tool.
Version Two: Covering the Full Platform
After the initial launch, AE Studio expanded Polaris to cover 100% of Dermsquared's public-facing content. The updated version introduced a query classification step that runs before any response is generated, routing each question through a specialised prompt strategy. This architecture means future capabilities can be added without disrupting what already works.
Helping Physicians Discover Content They Did Not Know to Look For
Beyond answering direct questions, Polaris surfaces related videos, podcasts, research, and conference materials after each response. Physicians stay connected to the broader Dermsquared knowledge base and find content they would not have searched for directly. This increases time on the platform and reinforces Dermsquared's value as the leading professional resource in dermatology.
OUTCOMES
What shipped.
Launched publicly March 2024 following staged rollout with pilot user testing
Content library spans 20+ years of expert dermatology content across articles, videos, podcasts, CME materials, case studies, and drug documentation
Query classification routes requests to specialized knowledge bases in milliseconds
Automated pipeline continuously indexes new content without manual intervention
Expanded in v2 to cover 100% of Dermsquared's public-facing website content
Powers the platform content recommender on video and podcast pages
KEY TAKEAWAYS
What we learned.
- Clinical AI requires specialized knowledge grounding. General-purpose search engines surface unvetted results. Polaris was built specifically on Dermsquared's verified expert content, giving physicians confidence that responses reflect trusted clinical sources rather than the open web.
- Query classification before response generation improves precision. Routing clinical, practice management, and institutional questions to category-specific prompts produces more accurate, contextually appropriate answers than a single generalized prompt strategy.
- Hybrid search preserves depth while adding speed. Combining AI-synthesized responses with full keyword search results gives physicians both an immediate answer and access to the source materials they may want to explore further.
- Automated content pipelines are essential for clinical relevance. Medical knowledge evolves constantly. An ingestion pipeline that automatically indexes new content ensures that Polaris reflects current clinical guidance without manual curation effort.
- Clinical accuracy guardrails are non-negotiable. Multi-layer verification and graceful handling of out-of-scope queries prevent hallucinations, which would erode physician trust and undermine the platform's clinical credibility.
- Phased architecture enables sustainable expansion. Building Polaris with a modular prompt strategy in v2 allowed AE Studio to expand the knowledge base and add new content types without disrupting the existing feature set.
IN SUMMARY
Bottom line.
In summary, Polaris transformed Dermsquared's content library from a resource physicians had to browse into one they could simply ask. As a result, by combining AI-augmented search with clinical accuracy guardrails, automated content ingestion, and intelligent query routing, AE Studio delivered a tool that fits naturally into the dermatologist's workflow.
The platform launched on time, scaled from a pilot cohort to the full Dermsquared physician community, and evolved through multiple versions to cover the entire knowledge base. Furthermore, Polaris now functions as both a search engine and a content discovery layer, surfacing related expert content alongside every clinical answer and deepening physician engagement with Dermsquared's 20+ year library.
For a platform built on the depth and authority of its expert content, Polaris made that depth accessible in real time, at the moment physicians need it most.
FAQ
Frequently asked.
How does Polaris ensure that clinical answers are accurate and not generated by general-purpose AI?
What types of clinical questions can Polaris answer?
How does Polaris stay current with new clinical information?
What is the difference between Polaris and a standard keyword search?
How does Polaris integrate with the broader Dermsquared platform?
LET'S TALK
Bring us the hard problem.
We'll bring the team that ships.