SigOpt

What SigOpt does

SigOpt, founded in 2014 and headquartered in San Francisco, builds an intelligent experimentation platform for hyperparameter optimization in machine learning and scientific computing. Its core engine runs Bayesian and other global optimization routines that explore multiple competing metrics under constraints, with native integration into XGBoost workflows and an open Core Module API for programmatic experiment management. The product portfolio consists of the main SigOpt platform, a self-hosted SigOpt-Server for teams that require data to remain on-premise, and SigOpt-Lite, a pip-installable in-memory version of the core module.

The customer base is anchored in academic and scientific research — published users include the University of Pittsburgh (additive and nanomanufacturing), MIT (machine-learning-driven materials discovery), and the University of Arizona (expensive lab experiments and physics-based simulations) — alongside machine learning practitioners who need automated hyperparameter tuning. The go-to-market is product-led and developer-centric: distribution is via GitHub, pip, and technical documentation, with a higher-touch self-hosted deployment offered for privacy-conscious enterprise teams.

SigOpt's business model is a freemium/open-core structure. The core platform and SigOpt-Lite are released as open source under a permissive model that drives adoption and community engagement, while revenue is implied to come from enterprise self-hosted deployments via subscription licensing, support, and customization. Backed by Andreessen Horowitz, DCVC, Blumberg Capital, Stanford, SV Angel, and Y Combinator, with a 2018 strategic investment from IQT, the company has disclosed roughly $8.7M in equity funding and operates with 11–50 employees. In 2024, SigOpt released its self-hosted server and in-memory module as open source, a move that broadens reach but likely constrains direct license revenue.

Name

SigOpt

Legal name

SigOpt

Website

https://sigopt.com/

Company type

Private

Founded year

2014

Operating status

Operating

Headcount range

11–50 employees

Short description

SigOpt provides a Bayesian optimization platform that automates hyperparameter tuning for machine learning and scientific computing. It serves ML practitioners and academic researchers, notably at MIT and University of Pittsburgh, via open-source tooling and an enterprise self-hosted deployment.

Ownership category

akta.pro rank

Product category

Machine Learning Optimization Software

NAICS

Computer Systems Design and Related Services (54151), Other Computer Related Services (541519)

SIC

Services-Computer Programming Services (7371), Services-Prepackaged Software (7372)

akta.pro primary industry

Fine-Tuning, Adaptation & Custom Model Training (PEFT/LoRA/RLHF) (HDAAACAC)

Keywords

• Hyperparameter optimization
• Machine learning tuning
• Bayesian optimization
• ML experimentation platform
• Model parameter tuning

Headquarters

HQ city

San Francisco

HQ country

United States

HQ region

North America

Markets served

GTM type

B2B

Offering type

Software

Cost components

Personnel, Technology or R&D, Marketing or Sales, Infrastructure, Operations

Revenue model

1. Enterprise Self-Hosted Solutions: SigOpt offers self-hosted server solutions for enterprise teams requiring data privacy and custom deployment environments. Revenue likely derived from enterprise licensing, support, and customization services for the self-hosted deployment option.
2. Open Source Platform (Free): Core platform released as open source with GitHub distribution - free for community use. This drives adoption and community engagement which may lead to enterprise upgrades.

Go-to-market motion1 record

Distribution channels3 records

Marketing channels4 records

Core offering

SigOpt provides an intelligent experimentation API and platform that tunes machine learning model parameters using Bayesian optimization. Its core platform is open-sourced with three delivery modes: a cloud-connected Core Module API, a self-hosted SigOpt-Server for privacy-sensitive enterprise environments, and SigOpt-Lite, an in-memory core module runnable locally via pip. The platform supports multi-metric and constraint-based optimization, experiment visualization, and native XGBoost integration for ML practitioners and scientific researchers.

Product overview

SigOpt is an intelligent experimentation platform that provides hyperparameter optimization for machine learning. The product portfolio consists of the core SigOpt platform (open source), the self-hosted SigOpt-Server deployment option for enterprise privacy, and SigOpt-Lite for lightweight in-memory local computation. All products share the core optimization engine accessible via the Core Module API, with XGBoost integration for seamless ML workflow support.

Differentiator

Problem solved

Functional benefit

Products and services

• SigOpt Core Platform The main SigOpt open-source intelligent experimentation platform that uses Bayesian optimization to tune machine learning model parameters. It explores multiple competing metrics, supports search-style experiments with many constraints, and provides experiment visualization and organization capabilities. Targeted at ML engineers, data scientists, and scientific researchers who need to automate and accelerate hyperparameter optimization.
• SigOpt-Server Self-hosted server deployment of SigOpt that allows teams to run the platform in their own environment so no data leaves their servers. Designed with customer privacy in mind and intended for enterprise teams and research groups that require data control and custom deployment. Available via GitHub (github.com/sigopt/sigopt-server).
• SigOpt-Lite A lightweight in-memory version of SigOpt's core module installable via pip ('pip install sigopt[lite]') that runs SigOpt computation locally while using the standard SigOpt connection interface to power a SigOpt experiment. Targeted at individual ML practitioners and researchers who want low-friction local experimentation.
• SigOpt Core Module API Public API endpoints that power SigOpt experiments, allowing developers to programmatically create and manage experiments, log parameters and metrics, and retrieve optimization results. Documented at docs.sigopt.com/core-module-api-references/api-endpoints and underlying the SigOpt-Server and SigOpt-Lite offerings.

Quantifiable outcome

• SigOpt helps researchers accelerate design of manufacturing processes through advanced math and optimization technology

Named customers3 records

Segments3 records

Ideal customer profiles3 records

Technology focussed Yes

API detail

Has API

Yes

API docs

API detail

Core technology

AI maturity

App detail

Integration1 record

AI capability2 records

Feature5 records

Partnerships

One partnership is on record.

• XGBoostcoreTechnology or IntegrationXGBoost users can more easily and efficiently learn hyperparameters for their models using SigOpt's integration. This integration allows XGBoost practitioners to leverage SigOpt's optimization capabilities directly within their XGBoost workflows.

Scale indicators1 record

Recent moves6 records

Expansion highlights4 records

Direct peers

• Weights & Biases: MLOps platform offering experiment tracking, dataset versioning, and a Sweeps product for hyperparameter optimization. Most direct commercial competitor with overlapping developer-led adoption motion and HPO as a core capability.
• Optuna: Open-source hyperparameter optimization framework from Preferred Networks with define-by-run API and strong academic adoption. Directly competes with SigOpt in the open-source HPO category with comparable Python-native developer experience.
• Determined AI: Open-source deep learning training platform with built-in hyperparameter search, experiment tracking, and distributed training. Comparable to SigOpt in targeting ML practitioners who need intelligent experiment management at scale.
• Comet: ML experiment tracking and model evaluation platform with hyperparameter optimization capabilities. Competes with SigOpt for the same ML practitioner persona and offers comparable experiment management and visualization.
• MLflow: Open-source ML lifecycle management platform from Databricks covering tracking, projects, models, and registry. Overlaps with SigOpt in the experiment tracking and management category, and serves the same developer-led adoption motion.
• Ray Tune: Distributed hyperparameter tuning library built on the Ray framework from Anyscale. Provides scalable HPO at the cluster level, directly competing with SigOpt for ML engineers running many concurrent tuning experiments.
• Hyperopt: Open-source Python library for serial and parallel optimization over awkward search spaces, originally from James Bergstra. Direct open-source competitor in the HPO category with academic and research community roots similar to SigOpt.

Broad incumbents

• Amazon SageMaker: AWS's broad ML platform includes built-in automatic model tuning (hyperparameter optimization) as one capability among many. A larger incumbent that offers overlapping HPO functionality as part of a much wider managed ML portfolio, often pre-selected by enterprise buyers already standardized on AWS.
• Google Vertex AI: Google Cloud's unified ML platform with Vertex AI Vizier for hyperparameter tuning. Competes as part of a full-stack cloud ML suite rather than as a standalone HPO specialist, making it a default choice for GCP-centric enterprises.
• DataRobot: Enterprise AutoML platform that automates model building and tuning across many algorithms. A broader incumbent that addresses the same underlying customer need (automated ML experimentation) but with a wider feature set and enterprise sales motion.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat4 records

Key highlights6 records

Customer concentration

• GitHubgithub.com/sigopt/sigopt-server
• Xx.com/sigopt
• Facebookfacebook.com/sigopt
• LinkedInlinkedin.com/company/sigopt

Revenue estimate

Valuation estimate

Number of profiles

Profiles3 records

Funding overview

Funding rounds4 records

Investors8 records

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about SigOpt