It imports your agent config from any supported platform into a universal IR (AgentGraph), then runs autonomous multi-turn conversation simulations against it. A simulator LLM plays the caller, your agent graph handles the routing, and an LLM judge scores transcripts against success criteria. Also supports deterministic rule tests for compliance stuff, PII leakage, required disclosures, forbidden phrases.

Write tests once and they work across platforms. Import from Retell, export to VAPI, run the same test suite. Also does format conversion between platforms if you're migrating.

Has interfaces for CLI (CI/CD), web UI, REST API, and a TUI. Results go into DuckDB so you can query them. Uses LiteLLM via DSPy so it works w/whatever provider you want

https://github.com/voicetestdev/voicetest

voicetest is an open source (Apache 2.0) test harness that works across voice AI platforms. You import your agent graph from any supported platform (or define one from scratch), write test scenarios with expected behaviors, and voicetest simulates conversations and evaluates them with LLM judges that score each turn 0.0-1.0 with written reasoning. It also ships global compliance evaluators for things like HIPAA, PCI-DSS, and brand voice consistency. The core abstraction is an AgentGraph IR that normalizes across platform formats, so you can convert between Retell, VAPI, LiveKit, and Bland configs and test them all the same way.

Quick start:

``` uv tool install voicetest voicetest demo --serve ```

That gives you a web UI at localhost with a sample agent, test cases, and evaluation results you can poke at. There's also a CLI, a TUI, and a REST API. It integrates into CI/CD with GitHub Actions, uses DuckDB for persistence, and includes a Docker Compose dev environment with LiveKit, Whisper STT, and Kokoro TTS. If you have a Claude Code subscription, voicetest can pass through to it instead of requiring separate API keys for evaluation.

GitHub: https://github.com/voicetestdev/voicetest Docs: https://voicetest.dev API reference: https://voicetest.dev/api/

Comments URL: https://news.ycombinator.com/item?id=47048811

Points: 3

# Comments: 0

https://github.com/voicetestdev/voicetest

- test harness for voice agents. - multi platform formats (Retell, VAPI, Bland, LiveKit) compile down to a unified AgentGraph IR - import from one platform, test locally, export to another - use litellm, DSPY to config models, if on a subscription use claudecode as a runner to avoid API call charges - metric judges produce continuous 0-1 scores instead of binary pass/fail since a 0.65 and a 0.35 both fail a 0.7 threshold but represent very different agent behaviors. - persist to DuckDB for querying across test history - adding auto-healing graph mutations where failed tests propose structural + prompt changes to the agent graph and validate against a regression suite

Wrote up the architecture here https://peet.ldee.org/general/2026/02/03/testing-voice-ai-ag...

The problem: I'd spin up 3-4 agents working on different features simultaneously, then conflict on files, and resolving those conflicts burns agent context. Git worktrees solve the isolation problem but the native CLI is verbose, lacks primitives for managing multiple sessions, and I'd have to manage persistence (folders to store the trees) separately.

wt wraps git worktree in an interface designed for this workflow:

  wt new feature/auth    # creates worktree + spawns subshell
  (wt:feature/auth) $ claude
  exit
  git merge feature/auth

There's a /do skill for Claude Code that implements issue-driven workflows: /do gh 123 fetches the GitHub issue, creates a worktree with a branch derived from the issue, and populates the agent context with the description.

Written in Rust. Binaries for macOS/Linux.

https://github.com/pld/wt

Blog post with more detail: https://peet.ldee.org/general/2026/01/26/wt-git-worktree-orc...

Comments URL: https://news.ycombinator.com/item?id=46765489

Points: 3

# Comments: 2

Comments URL: https://news.ycombinator.com/item?id=28205602

Points: 2

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=26228840

Points: 85

# Comments: 94

Comments URL: https://news.ycombinator.com/item?id=26013560

Points: 13

# Comments: 16

Our global health platform is used on smartphones by community health workers around the world to track and monitor their patients offline and on. Our data and ML platforms analyze patient risk, coordinate humanitarian projects, and manage Covid-19 tracking and messaging campaigns. We were named one of the “50 Most Genius Companies” by TIME magazine and have partnerships with the World Health Organization, the Bill and Melinda Gates Foundation, UNICEF, the United Nations Development Program, the World Food Program, and others.

References: https://blog.ona.io/news/2018/11/16/path-bid-story.html, https://blog.ona.io/canopy/covid19/2020/09/01/how-we-launche...

Our Global Health team stack is Android / Java / Postgres / React

Our Data Platform team stack is Django / Clojure / ClojureScript React (om -> reagent)

Our ML team stack is Beam / NiFi / Kafka / SciPy + Jupyter (prototyping) / Superset

We have full-time positions available, as well as fellowships and other opportunities.

https://ona.io/jobs.html

jobs+hn@ona.io

Comments URL: https://news.ycombinator.com/item?id=24270301

Points: 5

# Comments: 5

Comments URL: https://news.ycombinator.com/item?id=23114979

Points: 1

# Comments: 0

Named one of the “50 Most Genius Companies” by TIME magazine, we are a social enterprise building technology products that improve access to essential, life-saving services. We are funded by the World Health Organization, the Bill and Melinda Gates Foundation, UNICEF, the United Nations Development Program, the World Food Program, and others.

Ona's GIT fellowship is a 12-week, Nairobi-based engagement pairing experienced software engineers with a global health or humanitarian open source project.

Our Global Health mobile team stack is Android / Java / Postgres / React

Our Data Platform team stack is Django / Clojure / ClojureScript React (om -> reagent)

Our ML team stack is NiFi / Kafka / SciPy + Jupyter (prototyping) / Superset

We have full-time positions available, as well as fellowships and other opportunities to contribute, https://ona.io/jobs.html

Comments URL: https://news.ycombinator.com/item?id=21503929

Points: 1

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=20517031

Points: 2

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=20237530

Points: 2

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=19687790

Points: 2

# Comments: 0

We're looking for developers who want to build foundational data systems that drive change. Our team has worked on projects that record the social infrastructure of entire countries, tally the winners of national elections, reduce infant mortality, and decrease the spread of malaria. We build software that solves real problems and you will too. We were named a TIME Magazine Genius Company of 2018.

Global health stack

* Android, Java, Postgres

Data platform stack

* Python, Clojure, ClojureScript, PostGIS

Data warehouse stack

* NiFi, Kafka, Druid, Superset, React, Redux

Learn about our recent projects and what it's like to work at Ona, https://blog.ona.io/

Join us by applying here, https://ona.io/jobs/sw_eng.html Additional roles listed here, https://ona.io/jobs.html

Comments URL: https://news.ycombinator.com/item?id=18704258

Points: 3

# Comments: 0

We're looking for developers who want to build foundational data systems that drive change. Our team has worked on projects that record the social infrastructure of entire countries, tally the winners of national elections, and reduce infant mortality. We build software that solves real problems and you will too.

Global health stack

* Android, Java, Postgres

Data platform stack

* Python, Clojure, ClojureScript, PostGIS

Data warehouse stack

* NiFi, Kafka, Druid, Superset, React, Redux

Learn about our recent projects and what it's like to work at Ona, https://blog.ona.io/

Join us by applying here, https://ona.io/jobs/sw_eng.html

Comments URL: https://news.ycombinator.com/item?id=17794197

Points: 3

# Comments: 0