iamwilhelm

Meet Scout. The open-source company brain
YC's Summer 2026 Requests for Startups
 named two ideas that point at the same thing from different angles.
Company Brain
: pull knowledge out of fragmented sources (Slack, email, tickets), structure it, keep it current, turn it into something AI can act on.

    "Structure it, keep it current" is the wrong approach. 
    The trick is "navigation over search". More on this later.

Square profile picture

iamwilhelm

The Harness Is the Backend 
The most important architectural question in AI infrastructure right now isn’t which model to use. It’s how much infrastructure is required to build something useful with it.
Anthropic, OpenAI, CrewAI, LangChain all call that wrapping the agent harness. The harness includes the orchestration loop, tools (MCP, A2A), memory, context management, and error handling that make a model useful. They all agree the model isn’t the product. The infrastructure is. They disagree deeply on how much of it should exist.
Anthropic keeps their harness thin. It’s an elegant loop: Assemble the prompt, call the model, execute tool calls, and repeat. The model decides everything. OpenAI adds more structure: instruction stacks, orchestration modes, and explicit handoff patterns. CrewAI takes a multi-pronged approach: deterministic Flows for routing and validation, autonomous agents for the rest. LangGraph has the biggest harness Every decision is a node, every transition a defined edge, the entire workfl

iamwilhelm

-- Lean v4.23.0-rc2

import Std

open Std

variable {α : Type} [DecidableEq α]

/- ------------------------------------------------------------
   0) Your minimal abelian group structure (unchanged)

iamwilhelm

variable {α : Type} [DecidableEq α]

-- Define our own minimal abelian group structure (no mathlib)
structure AbelianGroup (G : Type) where
  add : G → G → G
  zero : G
  neg : G → G
  add_assoc : ∀ a b c : G, add (add a b) c = add a (add b c)
  add_zero : ∀ a : G, add a zero = a
  zero_add : ∀ a : G, add zero a = a

iamwilhelm

import gradio as gr
from openai import OpenAI
import jinja2
from transformers import AutoTokenizer

# get an api key from hyperbolic.
api_key = "..."

# Initialize the OpenAI client
client = OpenAI(

iamwilhelm

theory balancing
  imports Main
begin

type_synonym nodesizes = "nat list"

fun insert :: "nodesizes ⇒ nat ⇒ nodesizes" where
  "insert [] _ = []" |
  "insert (x#xs) 0 = (x+1)#xs" |
  "insert (x#xs) n = x # insert xs (n-1)"

iamwilhelm

theory lattice_merkle_tree
  imports Main
begin

fun opt_last :: "'a list ⇒ 'a option" where
  "opt_last [] = None" |
  "opt_last xs = Some (last xs)"


(****** Lattice Merke Tree ******)

iamwilhelm

// tic tac toe
import * as readline from "readline";

enum Turn {
  Player = "player",
  Computer = "computer",
}

type Piece = string;

iamwilhelm

import * as fs from "fs/promises";
import Knex from "knex";
const Web3 = require("web3");

import { Refract, Component, Fiber } from "./refract";

/******* custom hooks ******/

export const useFetchJson = filePath => {
  let [json, setJson] = Refract.useState("useFetchJson", null);

iamwilhelm

import Web3 from "web3";
import Knex from "knex";

import {
  loadTokenList,
  marketContract,
  borrowAndSupplyRate,
  rateToApy
} from "./compound.js";
import { subscribeNewBlocks } from "../libs/eth.js";