Fathom Tech Stack

AI medical coding automation for health systems

Hospitals and Health Care San Francisco, California 51–200 employees Founded 2017 Privately Held

Fathom automates medical coding using deep learning and NLP applied to clinical data. The stack reveals a data-heavy, cloud-native architecture (BigQuery, Redshift, Snowflake, GCP, Kubernetes) built for scale—a fit for processing thousands of patient encounters. Current project focus on Kubernetes operators and prompt engineering suggests the team is moving beyond static models toward dynamic AI feature tuning, while pain points around scaling backend systems and cost efficiency indicate they're hitting infrastructure limits typical of high-volume clinical AI workloads.

Tech Stack 25 technologies

Core StackZoom HubSpot Python BigQuery Redshift Snowflake Kubernetes Go Linux Docker Slack Prometheus ClickHouse ArgoCD GitHub Actions LinkedIn TikTok Instagram R SQL GCP Notion Loom

AdoptingKubernetes

What Fathom Is Building

◆Challenges

Overhead of meetings
Scaling engineering team
Lack of engaged community
Time-consuming note-taking
Improving reliability of mission-critical services
Scaling services for spiky workloads
Reducing operational costs
Scaling backend systems
Scaling distributed systems
Cost-efficiency

▲Active Projects

Cloud storage caching layer
Social media strategy
Content calendar management
Upgrade existing features with prompt engineering
Prototype new ai features
Fine-tune models for specific needs
K8s operator development
Backend services architecture evolution
Building a k8s operator
Build product marketing function

Hiring Activity

Decelerating10 roles · 2 in 30d

Department

Engineering

Marketing

Data

Sales

Seniority

Mid

Senior

Junior

Manager

Company intelligence

Find more companies like Fathom by tech stack, pain points and active projects

Get started free

About Fathom

Fathom applies deep learning and natural language processing to automate medical coding—the labor-intensive task of translating clinical notes into billing and regulatory codes. The company sells to health systems and revenue cycle teams managing high-volume coding backlogs. Founded in 2017 and based in San Francisco, Fathom operates as a 51–200 person organization with an engineering-heavy workforce building cloud infrastructure and ML models. The technical footprint spans GCP, Kubernetes, Python, and multiple data warehouses, reflecting the computational demands of clinical NLP at scale.

HeadquartersSan Francisco, California

Company Size51–200 employees

Founded2017

Hiring MarketsUnited States

Frequently Asked Questions

What tech stack does Fathom use?

Python, R, SQL, BigQuery, Redshift, Snowflake, Kubernetes, GCP, Docker, TensorFlow, Go, and supporting tools like GitHub Actions, ArgoCD, and Prometheus.

What is Fathom working on?

Cloud infrastructure (Kubernetes operators, backend architecture), prompt engineering for existing models, fine-tuning models for specialty-specific coding, and product marketing function development.