Your Codebase as a Graph You Can Ask
Two open-source projects turning repos and agent memory into searchable, self-hosted knowledge graphs.
CloudCast
Every week we share what we're building, distilled research on frontier tech, the hottest open-source tooling, and how you can use them to level up your cloud governance (FinOps) game; subscribe to stay up to date.
Live Last read · last published · next check
Latest episodes
Primitives & Guardrails for Agentic Systems
Attribution, chargeback, and budget guardrails for multi-step LLM agent systems in production.
Prompt -> Loop -> Graph -> ?
probabilistic systems to with deterministic outcomes.
Bedrock's prompt optimizer cuts model costs
AWS's new Advanced Prompt Optimization tool rewrites prompts across 5 models so you pay for what actually works.
AI Agents That Write Their Own Research
How autonomous paper-generation frameworks like AI Scientist and autoresearch are changing what 'human-in-the-loop' actually means.
"Good Vibes" VS "Bad Vibes"
A trustworthy health score for active AI coding sessions so you can codify that "feeling" every AI-pair-programmer knows all to well...
Stop Querying Your Own Memory
Open-source tools that let agents recall context across sessions without writing retrieval boilerplate from scratch.
There Are No Cost Centers
Every business unit is a term in the revenue equation — including the ones you've been told to minimize. Here's the math, and what it changes about where you spend.
Autonomous Agent Loops Cut Coding Toil
How Ralph and OpenHands run PRD items to completion without human hand-holding between steps.
Shell Scripts That Cut AWS Bills
How hand-rolled CLI automation from the swoodford/aws collection compares to modern FinOps tooling for SREs and architects.
Buy the Table Stakes, Build the Machine
Why transparency doesn't kill margin, why "want" is the real currency, and why the durable edge is no longer a product — it's the R&D shop that makes them
The path to $100M, Part 2 — the machine that runs sales
The orchestrator: deterministic deal classification, a scoring model, a state machine that mints its own next move, and the four safety nets that guarantee no deal ever goes dark
$100m arr with 2 employees: Part I
How we're building and monitoring systems to automate every aspect of our business — and hit $100M ARR with two people.
70% fewer tokens, zero config
AST-based code search, persistent agent memory, and usage coaching: the tooling stack your AI coding workflow is missing.
AoE2 villagers, text compression, and what actually makes a chatbot smart
Two recent AI papers — one a clever joke about Age of Empires II, the other a serious measurement of model skill — make a much sharper point together than either does alone.
Route LLM calls by cost, not just quality
How intelligent model routing cuts inference spend without sacrificing output accuracy for your workloads.
One CLI to Rule Your AI Agents
ASM centralizes skills across Claude, Cursor, Windsurf, and 10+ agents — no more scattered directories.
Graph Your Codebase, Shrink Your Bill
Pre-indexed knowledge graphs cut tool calls by 70% and keep context local—no model changes required.
The fourth input: why energy ties cloud FinOps together
Labor, capital, data — and energy. How grid simulation thinking reframes capacity planning and infrastructure cost models.
Cloudy, with a chance of rain.
2023-01-08