Why I picked this
Victor's instinct here cuts to the existential question nobody in AI tooling wants to say out loud: what if the orchestration layer just... evaporates? Swyx surfaces a fascinating tension from inside the model providers themselves — Anthropic's Claude Code team rewrites their harness every 3-4 weeks because they believe the model should do the heavy lifting, not the wrapper. That's not a technical preference, that's a philosophical position. And it puts every company building 'AI agent frameworks' in an awkward spot: you're betting your business on a layer that the people building the actual intelligence think shouldn't exist. The finance analogy is perfect — was it the trader's skill or the institutional seat? Except here, the seat is getting smarter every quarter, and the trader might be obsolete by Q3. This isn't just architecture debate, it's a market structure question with real consequences for where you place your bets.
Three lenses
If I'm building an AI product today, I'm watching this closely but not panicking yet — models still need guardrails, logging, fallback logic, and cost management that someone has to write. The question is whether that 'someone' is a $50M venture-backed framework company or just 200 lines of Python I maintain myself.
The Big Harness vs Big Model debate matters less to me than deployment reality — I need something my team can actually use Monday morning, and right now that's still a framework with docs and support. But if I'm evaluating a vendor whose entire value prop is 'orchestration,' I'm asking hard questions about their moat in 18 months.
Here's what nobody's saying: the framework companies know this is coming, which is why they're all pivoting to 'observability' and 'governance' — the last defensible layer before you're just reselling API credits. Watch for the rebrand wave in Q2.
“I'm not even sure these guys want me to exist - AI framework founder at OpenAI event”
Key takeaways
- Central debate emerging: 'Big Model' (minimal harness, model does everything) vs 'Big Harness' (orchestration/framework layer adds value) - mirrors finance debate about trader skill vs institutional position
- Model providers like Anthropic/OpenAI are philosophically minimalist on harness - Claude Code rewrites from scratch every 3-4 weeks, emphasizing 'thinnest possible wrapper' with all secret sauce in the model itself
- Existential threat to AI framework/orchestration companies as reasoning models improve - framework founders questioning their own necessity as models become more capable of self-orchestration
People mentioned
- Boris Cherny, Engineer @ Anthropic/Claude Code
- Cat Wu, Engineer @ Anthropic/Claude Code
- Ryan Lopopolo, Codex Team @ OpenAI
- Noam Brown, Researcher @ OpenAI
- Swyx, Author/Analyst @ Latent Space
Companies
Key metrics
- •rewritten from scratch every 3-4 weeks
- •$3M in profits (finance analogy)
Why this matters for operators: Critical for operators evaluating build-vs-buy decisions in AI agent infrastructure — reveals philosophical divide between model providers and orchestration vendors that will shape vendor viability over next 12-18 months
I cover AI×GTM intelligence like this every Wednesday.
Get STEEPWORKS WeeklyMore picks
Why It’s So Hard for Older B2B Leaders to Compete in AI: Your Customers Can Do A Lot in Claude for $20-$200/Month. And You’re Paying $1.00 Per API Call For the Good Stuff.
- B2B vendors face brutal unit economics: complex AI features cost $0.50-$2.25+ per API call while customers get unlimited access to same models for $20-200/month via Claude direct
- The pricing arbitrage creates existential threat to B2B AI wrappers - customers can increasingly do sophisticated analysis directly in Claude rather than through enterprise software
- Cheap AI features (pennies per customer) signal lack of competitive differentiation - genuinely valuable AI analysis requires expensive extended thinking modes and large context windows that compress margins
Confessions of an AI lab rat
- CEO-level AI adoption requires 1-2 hours daily commitment with disciplined feedback loops - casual usage produces unimpressive results that cause people to dismiss the technology prematurely
- Contrarian shift from 'subtraction story' (cost cuts/headcount reduction) to 'addition story' (3 new revenue lines economically impossible pre-AI) - suggests AI's bigger impact is enabling new business models rather than pure efficiency
- Critical deployment gap: AI capabilities exceed enterprise readiness due to security, system integration, and data access governance issues - agent-to-agent workflows exacerbate this problem at scale
Father of the iPod and iPhone on building taste, judgment, and creativity in the AI era | Tony Fadell
- Cognitive surrender to AI is the biggest risk facing product builders - legendary hardware/software creator warns against over-reliance on AI tools that erode taste and judgment
- Opinion-based decisions are essential for v1 products - data-driven approaches fail when building truly novel products (iPhone keyboard debate as case study)
- AI-generated code creates brittle, unmaintainable products - contrarian take from someone with 300+ patents on why AI coding tools may harm long-term product quality
This analysis was produced using the STEEPWORKS system — the same agents, skills, and knowledge architecture available in the GrowthOS package.