Companies

The Synthesis Company

thesynthesis.company

The Synthesis Company provides AI research assistants for fast, large-scale scientific evidence synthesis and systematic reviews.

HQSan Francisco, California, United States
Employees1-50
Jobs checked 3d ago
Specialization
Not classified
Business
Software Applications
Market
Life Sciencesand 1 more market values
  • Scientific Research
Ecosystem
Applications
Customers
Enterprisesand 2 more customers values
  • Governments
  • Researchers
More details
Organization type
Commercial Company
Ecosystem
Applications
Business
Software Applications
Specialization
Not classified
Market
Life Sciences, Scientific Research
Customers
Enterprises, Governments, Researchers
Delivery models
Hosted Software
Revenue models
Not classified
Listing status
Not classified
Latest equity round
Not classified
Reported funding amount
Not classified

About

The Synthesis Company builds AI-powered evidence-synthesis tools for systematic reviews, serving biopharma companies, researchers, and regulatory bodies. Its differentiation is automating large-scale literature analysis, enabling reviews of 20,000+ papers in weeks—up to 100x faster than traditional methods—with reported 98% accuracy.

Market

The company competes in AI-assisted scientific evidence synthesis and systematic literature review software, particularly for high-stakes pharmaceutical, academic, CRO, and AI-research applications. It positions Eva as a more rigorous and scalable alternative to conventional review processes and general academic-search tools by combining specialist agents for screening, extraction, and verification with human adjudication, continuously updated living reviews, and claimed performance of reviewing 20,000-plus papers in weeks with 98% accuracy.

Target Customers

The Synthesis Company targets biopharma and pharmaceutical companies, contract research organizations (CROs), universities and academic research groups, and AI labs that need rigorous, large-scale scientific evidence synthesis. Likely buyers are evidence-synthesis leaders, medical-affairs or research teams, CRO partners, academic investigators, and AI-research teams.

At a Glance

Problem

Scientific literature has become too large for researchers to synthesize manually, forcing them to narrow the questions they ask. The economics are especially painful in biopharma: evidence synthesis can take roughly 18 months and cost more than $141,000 per review, while incomplete evidence can distort billion-dollar decisions. The clearest killer use case is giving pharmaceutical companies complete evidence awareness quickly enough to identify safety signals, indication opportunities, and competitive developments before rivals do.

Product / Service

The Synthesis Company offers Eva, a team of specialist AI research agents for evidence synthesis. Eva screens literature, extracts relevant information, and verifies results, while human reviewers adjudicate the outputs. This is a human-in-the-loop software workflow rather than an autonomous replacement for expert review, and the company currently presents access through a demo-led model.

The claimed benefit is to make systematic reviews dramatically broader and faster: the company says it can handle reviews of more than 20,000 papers in weeks with 98% accuracy, compared with the conventional months- or years-long process. Its validation study covers 30,150 documents, and its broader vision also includes supplying AI labs with expert-validated scientific reasoning traces.

Market

The company competes in AI-assisted systematic literature review and scientific evidence-synthesis software, serving research-intensive organizations such as pharmaceutical companies, academic researchers, and AI labs. Relevant adjacent competitors include Elicit, which covers review protocol refinement, source gathering, screening, extraction, and synthesis; Rayyan, which focuses on screening, deduplication, and PICO extraction; and DistillerSR, which automates literature-based evidence synthesis for research and regulatory workflows.

The Synthesis Company appears to be an early-stage, validation-led business rather than one with publicly established revenue. Y Combinator lists it as an active Summer 2024 company founded in 2024, and the company reports collaboration with researchers at Stanford, Harvard, and Johns Hopkins, a 30,150-document validation study, and a capability claim of 20,000-plus-paper reviews in weeks. The public evidence shows a request-a-demo commercial posture, but does not disclose a verified customer count, revenue, or funding figure.

Founders & Leadership

Cyrus NourooziFounder
Cofounder & CEO
Amir MehrFounder
Cofounder & CTO

Funding History

2024-08
Seed / Y Combinator S24$500K (inferred YC standard deal)

Y Combinator

Recent News

2026-05-04product_launch
AI Verification Layer for CRO-Delivered SLRs

The Synthesis Company presented an AI verification layer for CRO-delivered systematic literature reviews, offering independent screening verification and white-label support. The page highlights 97.2% screening sensitivity and co-authored validation studies with CROs.

2025-11-14other
Why scientific evidence is the next frontier

The Synthesis Company published its thesis on scientific evidence synthesis, arguing that consumer AI tools lack audit trails, transparency, and expert validation. The company cites 98% accuracy on complex screening and data-extraction tasks.

Active Roles

0

No active roles right now.

Get notified when they post

Business Model

The company appears to use a demo-led B2B model, selling AI research-assistant and evidence-synthesis workflows to biopharma companies, CROs, researchers, and professors. Public materials do not disclose subscription pricing or a confirmed per-review rate.

Products

Eva — a multi-agent AI research assistant for systematic literature reviews and evidence synthesisLiving systematic review workflows that update as new evidence arrivesAI verification and validation layer for CRO-delivered systematic literature reviewsExpert-validated scientific reasoning traces and evidence-synthesis infrastructure for AI labs

Customers

Stanford (research collaborator; customer status not publicly disclosed)Harvard (research collaborator; customer status not publicly disclosed)Johns Hopkins (research collaborator; customer status not publicly disclosed)

Tech Stack

Large language models (LLMs)Multi-agent AI architecture using specialist research agentsDSPy-informed prompt/model orchestrationAI document screening, data extraction, and verification workflows

Competitors

Elicit
Consensus
DistillerSR
Covidence
Rayyan
EPPI-Reviewer