We re excited to soft-launch TowTurk, a smart roadside assistance marketplace designed to remove stress when drivers need help the most.
Instead of calling multiple providers, negotiating prices, or waiting blindly, TowTurk connects drivers with verified roadside assistance providers in real time with transparent pricing, live ETAs, and secure digital payments.
We open-sourced optimize_anything, an API that optimizes any text artifact. You provide a starting artifact (or just describe what you want) and an evaluator it handles the search.
import gepa.optimize_anything as oa result = oa.optimize_anything( seed_candidate="<your artifact>", evaluator=evaluate, # returns score + diagnostics
)
It extends GEPA (our state of the art prompt optimizer) to code, agent architectures, scheduling policies, and more. Two key ideas: (1) diagnostic feedback (stack traces, rendered images, profiler output) is a first-class API concept the LLM proposer reads to make targeted fixes, and (2) Pareto-efficient search across metrics preserves specialized strengths instead of
Yesterday was my 27th birthday. Instead of the usual dinner, I spent the morning in VIM (btw) shipping the waitlist for Squad In Sync.
The context: I was trying to plan my own birthday and NYE with my friends. Within 48 hours, the WhatsApp thread was 400+ messages deep. The flight info was buried, the restaurant link was lost, and half the group was in "decision paralysis" because they couldn't see the full plan.
