Sam Altman on Goldman Sachs: Why AI Coders Beat Finance

TL;DR: Sam Altman's recent comment that working at Goldman Sachs "sounds unbelievably terrible now" reflects a broader shift in developer priorities — from high-pressure finance toward building with AI. This article explores why that sentiment resonates, what it means for your career choices, and how you can start building with open-model APIs today.

Why Sam Altman's Goldman Sachs Comment Matters to Developers

Sam Altman's remark about Goldman Sachs isn't just celebrity gossip — it's a signal about how the tech industry's center of gravity has shifted. When the CEO of OpenAI says a prestigious finance job sounds "unbelievably terrible," he's speaking to a generation of developers who increasingly value building AI products over grinding through 80-hour weeks in banking.

The comment, made in a recent interview, highlights a real tension: traditional finance careers offer high pay but often involve repetitive tasks and rigid hierarchies. Meanwhile, AI development offers creative problem-solving, rapid iteration, and the chance to build tools that amplify your own productivity. For many developers, the choice isn't just about salary — it's about what kind of work feels meaningful.

This isn't about dismissing finance entirely. It's about recognizing that the skills you develop as an AI developer — prompt engineering, model evaluation, API integration — are increasingly valuable across every industry. And unlike the opaque world of investment banking, AI development is accessible: you can start building today with a free account and a few lines of code.

What Does "Open Models" Mean in the AI Landscape?

When we talk about open models in AI, we mean large language models (LLMs) whose weights and architectures are publicly available. Unlike proprietary models like GPT-4, open models like DeepSeek, GLM, and Qwen can be self-hosted, fine-tuned, and integrated without vendor lock-in.

For developers coming from a finance background — or anyone considering a pivot — open models offer several practical advantages:

• Cost transparency: You pay per token (input and output) with no hidden fees
• Flexibility: Switch between models based on your task's requirements
• No lock-in: Your code works with any OpenAI-compatible API, so you're never trapped with one vendor

Here's a quick comparison of popular open models available through TokShop's OpenAI-compatible API:

Model	Input (per 1M tokens)	Output (per 1M tokens)	Context Window
DeepSeek V3.2	$0.42	$0.63	128,000 tokens
GLM 4.6	$0.90	$3.30	200,000 tokens
Kimi K2	$0.855	$3.45	131,072 tokens
Qwen3 Coder	$2.25	$11.25	262,144 tokens

As you can see, there's a model for every budget and use case. DeepSeek is excellent for high-volume, cost-sensitive applications, while Qwen3 Coder's massive context window suits complex code generation tasks.

How Do You Start Building with Open Models?

Getting started with open models is straightforward — you don't need a finance degree or a Silicon Valley pedigree. Here's a practical path:

1. Create an account at TokShop with just your email and password
2. Generate an API key (looks like sk-tok-...) from your dashboard
3. Make your first API call using any OpenAI SDK

Here's a minimal Python example to get you started:

from openai import OpenAI

client = OpenAI(
    base_url="https://tokshop.xyz/v1",
    api_key="sk-tok-your-key-here"
)

response = client.chat.completions.create(
    model="deepseek-v3.2",
    messages=[
        {"role": "user", "content": "Write a Python function to calculate compound interest"}
    ]
)

print(response.choices[0].message.content)

The beauty of this approach is that you're learning transferable skills. The OpenAI-compatible interface means anything you build today works with other providers tomorrow. And with prepaid billing, you control exactly what you spend — no surprise invoices.

How Does AI Coding Compare to Finance Careers?

Let's be honest about the trade-offs. Working in finance can offer:

• Higher starting salaries at top firms
• Structured career progression with clear milestones
• Prestige and networking opportunities

But AI development offers:

• Immediate impact: Your code runs in production today
• Creative autonomy: You choose what to build and how
• Rapid skill growth: The field evolves weekly, not yearly
• Location independence: Remote work is the norm

The practical reality is that AI skills compound. A developer who learns to integrate models like GLM 4.6 or Kimi K2 into real applications gains experience that's directly transferable to any tech company. Meanwhile, finance skills are often specific to a single firm's systems and culture.

Should You Pivot to AI Development?

If you're considering a move from finance to AI development, here's a realistic assessment:

Do it if:

• You enjoy building things more than analyzing spreadsheets
• You're comfortable with ambiguity and rapid learning
• You want your work to have direct, visible impact

Think twice if:

• You value stability and clear career ladders above all
• You dislike writing code or debugging
• You prefer structured environments with established processes

The good news is that you don't have to make an all-or-nothing choice. Many developers start by building side projects with open models while keeping their day jobs. The pricing model at TokShop — pay-as-you-go with no subscriptions — makes experimentation affordable.

FAQ

Is it really "terrible" to work at Goldman Sachs?

No — Sam Altman's comment reflects his personal preferences, not an objective judgment. Goldman Sachs offers competitive compensation and valuable experience. The point is that for many developers, AI work offers different (not necessarily better) rewards: creative freedom, rapid iteration, and the ability to build products that reach millions.

What's the cheapest way to start experimenting with AI models?

DeepSeek V3.2 at $0.42 per million input tokens is currently the most cost-effective option on TokShop. You can start with a small prepaid credit and make hundreds of API calls for under a dollar. Check the pricing page for current rates.

Can I use open models for production applications?

Yes. TokShop's API is OpenAI-compatible, meaning you can integrate it into production systems using standard SDKs. The key considerations are model selection (match context window and cost to your use case) and monitoring your token usage to manage costs effectively.