Prev Next

AI / Core OpenAI Codex Application Fundamentals Interview Questions

What is the OpenAI token system and how do you count and optimise token usage?

OpenAI models process text as tokens - chunks of characters roughly 3-4 characters long for English text, or about 75% of a word. Pricing is per token (input + output), so understanding tokenisation directly impacts application costs.

Token counting rules of thumb
ContentApproximate token count
1 English word~1.3 tokens on average
1 page of text (~500 words)~650 tokens
1,000 characters~250 tokens
Short code function (20 lines)~80-150 tokens
Full file (200 lines of Python)~800-1500 tokens
import tiktoken

# Count tokens before sending (avoid surprises)
encoding = tiktoken.encoding_for_model("gpt-5.5")

def count_tokens(text: str, model: str = "gpt-5.5") -> int:
    enc = tiktoken.encoding_for_model(model)
    return len(enc.encode(text))

# Count tokens for a Chat Completions messages array:
def count_message_tokens(messages: list, model: str = "gpt-5.5") -> int:
    enc = tiktoken.encoding_for_model(model)
    total = 3  # reply overhead
    for msg in messages:
        total += 4  # per-message overhead
        for key, value in msg.items():
            total += len(enc.encode(str(value)))
    return total

# Example:
tokens = count_tokens("Write a Python function that sorts a list using quicksort.")
print(f"Prompt tokens: {tokens}")  # ~15 tokens

# Via API (most accurate, no tiktoken required):
response = client.responses.create(
    model="gpt-5.5",
    input="Explain recursion.",
)
print(f"Input tokens: {response.usage.input_tokens}")
print(f"Output tokens: {response.usage.output_tokens}")
print(f"Total: {response.usage.total_tokens}")

Cost optimisation strategies:

  • Use max_output_tokens to cap output length on tasks with known response sizes
  • Use prompt caching for repeated system prompts (40-80% better with Responses API)
  • Choose smaller models (gpt-5.4-mini, codex-mini-latest) for lightweight tasks
  • Use Batch API for non-real-time workloads (~50% discount)
  • Compress context: summarise long conversation histories instead of passing full history
What is the tiktoken library used for in OpenAI applications?
Which response field tells you how many tokens the model's output consumed?

Invest now in Acorns!!! 🚀 Join Acorns and get your $5 bonus!

Invest now in Acorns!!! 🚀
Join Acorns and get your $5 bonus!

Earn passively and while sleeping

Acorns is a micro-investing app that automatically invests your "spare change" from daily purchases into diversified, expert-built portfolios of ETFs. It is designed for beginners, allowing you to start investing with as little as $5. The service automates saving and investing. Disclosure: I may receive a referral bonus.

Invest now!!! Get Free equity stock (US, UK only)!

Use Robinhood app to invest in stocks. It is safe and secure. Use the Referral link to claim your free stock when you sign up!.

The Robinhood app makes it easy to trade stocks, crypto and more.


Webull! Receive free stock by signing up using the link: Webull signup.

More Related questions...

What is OpenAI Codex and how has it evolved from its original form? What are the current recommended models for OpenAI Codex and the API, and when do you choose each? What is the OpenAI Responses API and how does it differ from the Chat Completions API? What are the built-in tools available in the OpenAI Responses API? What is the OpenAI Agents SDK and what are its four core primitives? What is the Codex CLI and what are its key features? What is the Chat Completions API and when should you still use it? What is function calling (tool use) in the OpenAI API and how does it work? What are structured outputs in the OpenAI API and how do you use them? What is the OpenAI Assistants API and what is its deprecation timeline? What is prompt caching in the OpenAI API and how does it reduce costs? What are reasoning models in the OpenAI API and what is the 'effort' parameter? What are OpenAI's key API authentication and security concepts? What are rate limits in the OpenAI API and how do you handle them in production? What is the Batch API and when should you use it? What is streaming in the OpenAI API and how do you implement it? What is OpenAI's Codex Skills feature and what are Automations? What is Model Context Protocol (MCP) and how does it integrate with OpenAI tools? What is fine-tuning in the OpenAI API and when should you use it? What are embeddings in the OpenAI API and what are they used for? What is the OpenAI moderation API and why is it important for application safety? What are OpenAI's image generation models and how do you use them in applications? What is the OpenAI Realtime API and what use cases does it enable? What is retrieval-augmented generation (RAG) and how do you implement it with OpenAI? What is OpenAI's approach to responsible use and safety in the API? What is the OpenAI token system and how do you count and optimise token usage? What are guardrails in the context of OpenAI application development? What is the OpenAI Files API and how is it used for document management? How do you implement multi-agent systems using the OpenAI Agents SDK? What is the OpenAI Evals framework and why is evaluation critical for production applications? How do you implement error handling in OpenAI API applications? What is the OpenAI Codex App and what are its main features for software teams? What are the key differences between OpenAI's o-series reasoning models and the GPT series? How does the Codex IDE extension integrate with development environments? What are the key considerations for building production-grade OpenAI applications? What is the OpenAI Codex and API pricing model and how do you estimate costs? What is the OpenAI computer use capability and what does it enable? How do you use the OpenAI API for code generation, review, and debugging tasks? What is the role of system prompts (instructions) in OpenAI applications and how do you design them effectively? How do you handle context window management in long-running OpenAI applications?
Show more question and Answers...

Database

Comments & Discussions