Archive edition · Aug 1–15, 2026

Edition — August 1–15, 2026

6 standout picks across 14 categories. Backfilled per our research methodology: candidates sourced from Hacker News and GitHub within this fortnight's window, then checked against each product's own site or repository. Every pick carries a dated evidence link.

AI Compute & GPU Cloud
PC

Perimeter Compute

Distributed AI cloud in city office buildings

Distributed GPU cloud using office buildings' spare energy.

Perimeter Compute launched as a distributed AI cloud provider that deploys GPU compute inside commercial office buildings in major US cities, tapping their excess energy capacity instead of building new data centers. The distributed-edge approach aims to add GPU supply closer to demand while using power that would otherwise go unused, targeting AI teams that need additional capacity. (Reported by DataCenterDynamics; company scale not independently deep-verified here.)

Why it mattered

A novel answer to the GPU-supply crunch — compute placed in office buildings on otherwise-wasted energy — reported as an Aug 2026 launch by an industry outlet.

Evidence dated 2026-08-04news ↗
Researched · pending approvalVisit site ↗
AI Model Serving & Inference
No verified pick this edition

No credible in-window candidate met the evidence bar for this category this edition — we flag it rather than force a weak pick.

AI Data Infrastructure
P

pgbot

Postgres intelligence for AI agents

Natural-language Postgres intelligence layer for AI agents.

Why it mattered

Brings natural-language Postgres intelligence to AI agents, reflecting the push to make production databases first-class agent tools.

Evidence dated 2026-08-11github ↗
Researched · pending approvalVisit site ↗
AI Development & Orchestration
XS

Xyne Spaces

The AI Org-OS for humans and agents

Collaborative org workspace for humans and AI agents.

Why it mattered

An open 'AI Org-OS' positioning humans and agents as collaborators in one workspace — from the Juspay team, part of the agent-orchestration surface wave.

Evidence dated 2026-08-02github ↗
Researched · pending approvalVisit site ↗
AI Observability & Evaluation
No verified pick this edition

No credible in-window candidate met the evidence bar for this category this edition — we flag it rather than force a weak pick.

AI Coding & Developer Tools
F

Fuxi

Terminal AI coding agent, multi-provider

Single-binary terminal coding agent with multi-provider routing.

Fuxi is a single-binary terminal coding agent that routes across 10+ LLM providers (DeepSeek, OpenAI, Claude, Qwen and more) with automatic failover and provider racing for reliability. It ships a deep tool suite — 50+ capabilities spanning file editing, shell commands and code search — plus persistent sessions with long-term memory and MCP client support. Permission-gated autonomous execution and extensibility through hooks, skills, plugins and slash commands let even non-developers build by conversation, and it installs as a single static binary with no runtime dependencies.

Why it mattered

A single static binary that routes across 10+ LLM providers with automatic failover and MCP support — part of August 2026's wave of local, provider-agnostic terminal coding agents.

Evidence dated 2026-08-04github ↗official ↗
AI Agents
PH

Phone Harness

Let your AI agent control your phone

Agents tap, type and scroll real iOS/Android apps.

Phone Harness lets AI agents operate real iPhones and Android devices — tapping, typing and scrolling through apps exactly as a person would, using visual input and on-screen controls with no jailbreaking or special APIs. Teams can use hosted cloud phones or connect their own device, giving agents genuine mobile-app reach instead of brittle per-app integrations. It targets developers building agents that need to act inside everyday consumer mobile apps.

Why it mattered

Gives agents real tap/type/scroll control of iOS and Android without jailbreaking or special APIs — a concrete step toward agents operating everyday mobile apps.

Evidence dated 2026-08-07github ↗official ↗
AI Sales
No verified pick this edition

No credible in-window candidate met the evidence bar for this category this edition — we flag it rather than force a weak pick.

AI Voice Agents
No verified pick this edition

No credible in-window candidate met the evidence bar for this category this edition — we flag it rather than force a weak pick.

AI Evals
No verified pick this edition

No credible in-window candidate met the evidence bar for this category this edition — we flag it rather than force a weak pick.

AI Observability
No verified pick this edition

No credible in-window candidate met the evidence bar for this category this edition — we flag it rather than force a weak pick.

AI Healthcare
No verified pick this edition

No credible in-window candidate met the evidence bar for this category this edition — we flag it rather than force a weak pick.

AI Biotech & Drug Discovery
No verified pick this edition

No credible in-window candidate met the evidence bar for this category this edition — we flag it rather than force a weak pick.

AI Monetization & Billing Infrastructure
C

CostPerPrompt

Live AI API pricing & cost calculators

Real-time pricing and workload cost calculators for AI APIs.

CostPerPrompt tracks live pricing for hundreds of AI models across major providers, refreshed automatically from public provider listings. Beyond a raw price table, it offers purpose-built calculators that model real workloads — chatbots with growing context, RAG pipelines split into indexing/retrieval/generation, and voice AI billed across STT + LLM + TTS — plus GPU-rental and image-API pricing and a token counter. It's a reference tool for teams estimating and controlling usage-based AI spend before it lands on the bill.

Why it mattered

Tracks live pricing across hundreds of models with workload calculators (chatbot, RAG, voice) — the cost tooling that matters as teams move to usage-based AI spend.

Evidence dated 2026-08-02hackernews ↗official ↗