A Day in the Life of a Data Scientist in 2026

2026年数据科学家的日常

How AI has massively changed my day-to-day workflow 人工智能如何彻底改变了我的日常工作流程

Haden Pelletier | Aug 14, 2026 | 5 min read Haden Pelletier | 2026年8月14日 | 阅读需5分钟

Two Years Ago, My Day Looked Completely Different

两年前，我的工作状态截然不同

Believe it or not, two years ago, I was still writing and debugging code daily. Line by line. Practically slamming my head against the desk after 2 hours of debugging to no avail. I know, it sounds kind of crazy, right? 信不信由你，两年前我每天还在写代码和调试代码。一行一行地写。在调试了两个小时却毫无进展后，我几乎想把头撞在桌子上。我知道，这听起来有点疯狂，对吧？

A normal day for me: 我的一天通常是这样的：

Writing (and debugging) every SQL query and Python script from scratch
从零开始编写（和调试）每一个 SQL 查询和 Python 脚本
Building slide decks bullet by bullet
一点一点地制作幻灯片
Writing documentation nobody would read until something broke (and even then, they hardly would)
编写没人会看的文档，直到系统出故障（即便那时，他们也几乎不看）

I wrote about my day in the life as a data scientist back in 2024. And almost none of that is what my actual Tuesday looks like now. I won’t pretend that’s purely a good thing. Some days it feels less like my job got easier and more like it quietly turned into a different job. One I had to learn on the fly, while I was still doing the old one. 我曾在 2024 年写过关于数据科学家日常的文章。而现在，我周二的实际工作内容几乎与那时完全不同。我不会假装这全是好事。有时我觉得工作并没有变轻松，反而像是悄然变成了一份完全不同的工作。我必须在做旧工作的同时，边学边做这份新工作。

Prompt Engineering Is a Major Part of the Job

提示词工程是工作的重要组成部分

Yes, we had ChatGPT in 2024. We had prompt engineering. But I was not doing it because ChatGPT often frustrated me. It was more work to explain the context behind what I was doing before feeding ChatGPT my code, and even then it would still not seem to be able to find the bug. 是的，我们在 2024 年就有 ChatGPT 了，也有提示词工程。但我当时并不怎么用，因为 ChatGPT 经常让我感到沮丧。在把代码喂给它之前，我得花更多精力去解释背景信息，即便如此，它似乎还是找不到 Bug。

Advancements in AI, especially those with Claude, have changed a lot of that attitude. Things like projects and skills have made it much easier to discuss your project with an AI that already knows the context and history behind it. So a meaningful chunk of my day now goes into writing and refining prompts. 人工智能的进步，尤其是 Claude 的进步，改变了这种态度。诸如“项目”和“技能”之类的功能，使得与已经了解项目背景和历史的 AI 进行交流变得容易多了。因此，我每天有相当一部分时间花在编写和优化提示词上。

Early on, my prompts were lazy. Something like: Summarize the forecast accuracy for this model. Which gets you a vague paragraph that often doesn’t contain the insights you really want. 起初，我的提示词很敷衍。比如：“总结该模型的预测准确性。”这只会得到一段模糊的文字，通常不包含你真正想要的见解。

Now I write prompts closer to: Summarize this model’s forecast accuracy over the last 14 days. Report the exact MAPE and RMSE for each day, flag any day where MAPE exceeded 5%, and state whether the trend is improving or degrading week-over-week. Do not round error metrics, report them to two decimal places. 现在我写的提示词更接近于：“总结该模型过去 14 天的预测准确性。报告每一天的精确 MAPE 和 RMSE，标记出 MAPE 超过 5% 的日期，并说明趋势是周环比改善还是恶化。不要对误差指标进行四舍五入，保留两位小数。”

The difference in output quality is enormous, and honestly, that’s now a skill I have to actively strengthen. 输出质量的差异是巨大的，老实说，这现在是我必须主动加强的一项技能。

A few things that are now part of my regular workflow: 现在我日常工作流程的一部分包括：

Double checking LLM model outputs
对大模型（LLM）的输出进行二次核对
Testing prompt variants against the same task and comparing outputs side by side
针对同一任务测试不同的提示词变体，并对输出进行对比
Writing constraints directly into the prompt (units, decimal precision, what not to guess at) instead of correcting the output after the fact
直接在提示词中写入约束条件（单位、小数精度、哪些内容不要猜测），而不是事后去修正输出

Finding Cost-Effective LLM Solutions (& Cutting Token Usage)

寻找高性价比的 LLM 方案（并减少 Token 使用量）

LLMs are expensive. Much more so than XGBoost models. This means that a lot more consideration needs to go into utilizing LLMs to analyze large datasets. The same data science principles still apply, though: 大模型很昂贵，远比 XGBoost 模型贵得多。这意味着在利用大模型分析大型数据集时，需要进行更多的考量。不过，数据科学的基本原则依然适用：

When a simpler heuristic or model can perform the task, always go with that first.
当简单的启发式方法或模型能完成任务时，优先选择它们。
Always clean your data before feeding it into a model. Garbage in=garbage out.
在将数据喂给模型之前，务必进行清洗。垃圾进，垃圾出。
Perform feature selection and select only meaningful features before training an ML model so you don’t shove hundreds of random features and cause overfitting or too much noise.
在训练机器学习模型之前进行特征选择，只保留有意义的特征，这样你就不会塞入数百个随机特征，从而导致过拟合或过多的噪声。

These pillars map over very well to LLMs. Not every task needs the biggest, most expensive model available. Routing the easy stuff to a smaller, cheaper model and reserving the expensive one for tasks that need it turned into a real cost lever. 这些支柱原则同样适用于大模型。并非每个任务都需要最强大、最昂贵的模型。将简单的任务分配给更小、更便宜的模型，而将昂贵的模型留给真正需要的任务，这已成为控制成本的有效手段。

Here are some examples of how I work on limiting costs: 以下是我如何控制成本的一些例子：

Data cleaning to cut down input sizes (for example, removing links, images, and other characters not relevant to the model from an email chain)
数据清洗以减少输入规模（例如，从邮件链中删除链接、图片和其他与模型无关的字符）
Caching repeated calls instead of re-running the same prompt against the same input
缓存重复的调用，而不是对相同的输入反复运行相同的提示词
Using traditional ML when appropriate instead of an LLM for everything
在适当的时候使用传统机器学习，而不是事事都用大模型
Tracking token spend per task
跟踪每个任务的 Token 消耗
Researching best practices for reduced token usage
研究减少 Token 使用量的最佳实践

Stakeholder Communication and Presentations

利益相关者沟通与演示

Here’s where a lot of saved time goes: meetings, slides, and translating what a model did into something a non-technical stakeholder can act on. I used to spend hours building a deck from scratch. Now I can generate a rough draft of a stakeholder-ready dashboard or slide outline in minutes. 这就是节省下来的大量时间去向：会议、幻灯片，以及将模型的工作成果转化为非技术背景的利益相关者可以采取行动的内容。我过去常花几个小时从零开始制作演示文稿。现在，我可以在几分钟内生成一份可供利益相关者使用的仪表板草稿或幻灯片大纲。

In practice, it just means I spend that freed-up time in more meetings, walking people through what the model found and why it matters, because the turnaround is fast enough that stakeholders expect check-ins more often. 实际上，这意味着我把节省下来的时间花在了更多的会议上，向人们解释模型发现了什么以及为什么这很重要，因为周转速度足够快，利益相关者期望更频繁地进行沟通。

The actual skill that matters here hasn’t changed: taking something technically true and making it something a product manager or executive can make a decision from. AI can draft the slide. It can’t decide what the point of the slide is (that’s still me.) 这里真正重要的技能并没有改变：将技术上正确的东西转化为产品经理或高管可以据此做出决策的内容。AI 可以起草幻灯片，但它无法决定幻灯片的重点是什么（那仍然是我要做的事）。

Conclusion

结论

Even with all of this, most of my job is still the same underneath. I still have meetings and need to collaborate with my team members. I still have to decide what’s worth modeling in the first place. I still have to catch when an AI-generated summary confidently states something that isn’t true. I still have to know the domain well enough to notice when a number looks slightly wrong instead of obviously wrong. And I still use traditional ML when necessary. 尽管如此，我的工作本质上大部分还是没变。我仍然需要开会，需要与团队成员协作。我仍然需要决定什么值得建模。我仍然需要识别 AI 生成的摘要中那些自信地陈述错误事实的地方。我仍然需要具备足够的领域知识，以便在数据看起来“略有偏差”而非“明显错误”时能察觉出来。而且在必要时，我依然会使用传统机器学习。

If anything, that judgment matters more now, not less because it’s the one part of the day that never got automated. My day in 2026 isn’t shorter than it was in 2024. It’s just shaped differently. Less time doing the mechanical parts, more time on the parts that require me to think deeper about business problems. 如果说有什么变化，那就是这种判断力现在变得更加重要，而不是更不重要，因为这是我一天中唯一没有被自动化的部分。我在 2026 年的一天并不比 2024 年更短，只是形态不同了。花在机械性工作上的时间少了，花在需要深入思考业务问题上的时间多了。