Hark Labs' model + agent + VM experience beats Codex, says early reviewer
Scobleizer · x · 2026-11-10
JJ Englert (retweeted by Scobleizer) says Hark Labs offers a rare world-class combination of model + agent + VM, arguing it beats OpenAI Codex, which he calls the closest competitor.
He highlights the polish of the experience, noting that solid browser use is easy to overlook but unlocks far more capability once an agent gets basic tasks right.
More from coding & agent
- Forgetful AI Agent Derails Enterprise Pilot: Rereading Context Ate 7% of Daily Usage — thedealdirector · 2026-11-10
- Ex-Goldman quants raise $1M from YC & Afore to build Volaren, a text-based AI investing agent — LinusEkenstam · 2026-11-10
- OpenAI launches Trips, a travel planning demo built on Responses API with web search — jxnlco · 2026-11-10
- Benchmarking 7 coding agents: memory varies 7x, idle CPU varies 2400x — apollohwk · 2026-11-10
- YC-backed 0.security says it found 30+ Linux kernel 0-days and 500+ CVEs — ycombinator · 2026-11-10
- flownav: read-only MCP server turns n8n workflows into a typed graph to save agent context — omert-dev · 2026-11-10