Tessl
AI-native software development platform built around specs and agent skills.
About
Tessl was founded by Guy Podjarny (founder of Snyk) and raised $125M at a reported $750M valuation to reimagine software development as spec-centric rather than code-centric: developers describe what to build in natural language specifications, and AI handles implementation and ongoing maintenance. In 2026 the company leaned into agent skills as the core unit — a platform for developing, evaluating, and distributing versioned, tested skills and context that keep AI coding agents correct as libraries and models change.
The honest caveat: Tessl spent a long time in waitlist and early access, so public evidence of production usage is thinner than the funding suggests. The spec-to-software vision is ambitious and unproven at scale, and teams deeply invested in existing IDE-based AI workflows may find the paradigm shift a hard sell.
What it can do
- Natural-language software specifications compiled to working code
- Automated detection and fixing of performance and security issues
- Versioned, tested agent skills as shareable dependencies
- Sandbox testing environment before deployment
- Multi-language support starting with Java, JavaScript, and Python
Setup
Join the waitlist or request access at tessl.com. The platform is web-based with integrations for existing developer tools.
Prerequisites
None beyond a modern browser; team onboarding for the skills platform.
Use cases
Works with
Trust score — 7/10
Scored 7/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 44 (tools-a); Trending score 72/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.