Eric Florenzano ericflo

Skipping the Cold Start: Agentic LLMs and the Next Social Web

TL;DR: Agentic LLMs change the launch calculus for social products. You no longer need a big user base to get compelling content and engagement. In fact, you can raise the bar on content quality—far beyond what humans will reliably do—by making agents meet strict requirements (even shipping runnable code) before anything gets posted. Think of it as moving from the Web‑2.0 era of optimizing for user‑generated content to a new era of eliciting the highest‑quality AI‑generated content.

The Old Playbook: Cold Starts and Crowd Drag

If you’ve ever tried to build a social or community product, you know the two classic headaches:

Start With the Judge: A Practical Blueprint for Dataset Engineering

TL;DR: Most LLM “training” work is really dataset engineering: defining the task, crafting a tiny set of crystalline examples, and building a reliable judge that can score outputs. If you start by perfecting the judge and then use it to drive generation, selection, and reinforcement learning—plus a few pragmatic guardrails—you can turn a tinkery, manual grind into a repeatable pipeline (and eventually, an automated agentic system).

The pattern today (and why it’s exhausting)

You notice a recurring failure: maybe the model keeps botching a class of SQL problems (“write a query that does X with window functions”), or it can’t follow a bespoke DSL, or it slips out of character in voice-constrained writing. The current playbook looks like this:

NO MOCKS, NO FALLBACKS, FAIL FAST, FAIL LOUD

This codebase follows a strict policy:

🚫 NO MOCKS

Mock objects create false confidence
They diverge from production behavior
They hide real integration issues
They make debugging harder

Movie Title: The Chalice of Light

Length:

2 hours and 45 minutes

Supporting Evidence -- Reason for Virality:

Many people believe in various kinds of supernatural phenomena for a variety of reasons, reasons I believe are not yet totally understood by academics, and certainly not by the general public. However, the fact that people believe in these things is by itself an untapped power.

	#!/usr/bin/env python3
	# /// script
	# requires-python = ">=3.11"
	# dependencies = [
	# "datasets",
	# "numpy",
	# "requests",
	# "tqdm",
	# "pyarrow"
	# ]

	def disk_cache(cache_dir: str = ".cache"):
	"""
	Decorator that implements disk caching for functions using JSON.
	NOTE: This is adapted for async usage: we await the wrapped function,
	but the file I/O remains synchronous for simplicity.
	Args:
	cache_dir: Directory to store cache files
	"""

	def decorator(func):

	import copy
	import json
	import os
	import logging
	import random
	import traceback
	from concurrent.futures import ThreadPoolExecutor, as_completed
	from dataclasses import dataclass

	import fire

	# create_pairs.py
	import argparse
	import copy
	import json
	import random
	from tqdm import tqdm
	from datasets import load_dataset

	from vllm import LLM, SamplingParams
	from transformers import AutoTokenizer

	# Deploy 1x AMD MI300X
	# python -m vllm.entrypoints.openai.api_server --port 8083 --host 127.0.0.1 --model meta-llama/Meta-Llama-3.1-70B-Instruct --max-model-len 120000
	# NUM_WORKERS=32 MODEL_NAME="meta-llama/Meta-Llama-3.1-70B-Instruct" OPENAI_API_URL="http://127.0.0.1:8083/v1" python agent_instruction_database.py

	import copy
	import os
	import json
	import traceback
	import random
	from pprint import pprint

	# MODEL_NAME="meta-llama/Meta-Llama-3.1-8B-Instruct" OPENAI_API_URL="http://localhost:1234/v1" OPENAI_API_TOKEN="..." python general_function_calling.py
	# MODEL_NAME="meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo" OPENAI_API_URL="https://api.together.xyz/v1" OPENAI_API_TOKEN="$TOGETHER_API_KEY" python general_function_calling.py
	# MODEL_NAME="meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo" OPENAI_API_URL="https://api.together.xyz/v1" OPENAI_API_TOKEN="$TOGETHER_API_KEY" python general_function_calling.py

	import copy
	import os
	import traceback
	import json
	import re
	import inspect