may be one of the crucial necessary phases of our careers.
I’m not saying this to be dramatic or clickbaity but because something subtle and irreversible is going on in the way in which I work. With each passing day, I find myself using AI more. I’m going less forwards and backwards with it. I query it less because with increasing exchange, it has turn into more often than not.
My role is slowly moving from to .
Today, I get used to watching AI handle things before I work on things that I once thought required my expertise.
I often joke that I might never use ChatGPT for planning my travels. Travel planning is my playground. I like opening twenty tabs, comparing neighborhoods, reading reviews, and constructing an itinerary that feels excellent. And yet, every week ago, I asked ChatGPT to walk me through all the pieces a first-timer at Disney Parks should know. In seconds, I had notes of all the pieces I should know and do, without opening some other tab.
That made me pause.
If AI can handle something I genuinely enjoy and take pride in… what does that mean for the remaining of my work?
My Workflow Before AI
Not way back, my work as an analytics consultant was long, nuanced and deeply tangible.
I might:
- Define the business problem
- Discover the correct data sources
- Write code from scratch to scrub messy datasets
- Manipulate and analyze the info
- Hit errors, debug for hours
- Search Stack Overflow, rewrite queries
- Explore edge cases
- Construct stakeholder decks
- Translate technical outputs into business narratives
A number of my value lived in executing this workflow.
Over time, I actually have worked to create a distinct segment for myself to have the opportunity to translate data for the business and vice versa.
What It Looks Like Now
Nonetheless, today, AI is commonly the very first thing that touches my problem statements.
Initially, I used to be mostly about experimenting with the prompts. I might describe the business context, the schema, the boundaries, and the expected consequence, and I explored what AI could do for me. Now that I actually have seen the productivity boost, the articulation of a few of my thoughts, I heavily depend on AI now to:
- Write end-to-end code for data cleansing, evaluation, and visualization
- Suggest features and improve model performance
- Surface insights I hadn’t considered
- Document your entire process
- Generate executive summaries for various audiences
With that, AI has effectively turn into my first analyst.
And this didn’t occur overnight and even in every week. The subtle shift happened over months and now, if I actually have something that should get done, I’m naturally inclined to go to AI first, even before I even fully think it through myself and I find that each exciting and deeply unsettling.
Because this shift isn’t incremental. It’s exponential.
I fear that we’re about to see AI replace multiple skill — coding, evaluation, writing, and more. It’s not only improving at one thing—it’s improving at , unexpectedly.
What This Really Means
AI is becoming a general layer for cognitive work.
I don’t know if AI will ever replicate deep human empathy or if trust built over years may be automated. And truthfully, I don’t know where the ceiling is anymore.
But I do have a sense that the individuals who will navigate this shift well will not be those avoiding it however the ones leaning into it with curiosity.
So Where Do We Construct an Edge?
I’ve been desirous about this rather a lot currently—when human intelligence gets normalized by artificial intelligence, how do I stay relevant? I don’t want to find yourself watching my role slowly reshape itself without me reshaping my skills and toolkit too.
I’ve realized that the sting is becoming less visible.
Up to now years, once I joined the workforce as an analyst, I believed that because I do know SQL, I can construct models, and I can clean messy data, I actually have an edge. These were tangible skills one could measure, improve, and showcase. Nonetheless, a variety of that’s slowly getting abstracted away. AI can do most of it fast, and increasingly well.
So the sting has to maneuver someplace else.
For me, it’s beginning to feel just like the edge is in how you’re thinking that before you even open a tool.
And here’s how I’m preparing to construct that edge for the following few years to come back as a senior analyst –
- Get hands-on with AI in your workflow:
I highly recommend beginning to use AI seriously (not only searching itineraries and cleansing up your emails). The sting comes from leveraging AI for , not passive usage.- Don’t stop at “write me a question” or like a search engine. Use it for full problem cycles from data cleansing to evaluation to storytelling with that data.
- Compare its output with yours and spot the gaps.
- Understand where AI works for you, and more importantly, where it doesn’t:
The true edge isn’t in only using AI. It’s knowing when to depend on it. AI can generate answers, but you have to know after they’re flawed.- At all times ask if the trend/pattern/insight that AI is suggesting is smart? What’s missing? What’s biased?
- Pressure-test outputs with easy sanity checks.
- Be intentional about what you delegate
Let AI handle speed, structure, and first drafts for now as I get settled on this space, if not already. Next, move as much as letting AI take care of problem framing, judgment, ethics, and accountability. But, don’t forget to validate.- Cross-check results with small samples, edge cases, or alternate queries.
- Don’t trust clean outputs blindly. At all times confirm those outputs.
- Prepare on your role to evolve.
We’re already moving from being query writers to prompt thinkers, data validators, and storytellers.- Transcend “here’s what the info says” → “here’s what we must always do next.”
- Tie evaluation to business impact, not only accuracy.
That is where analysts start becoming decision partners - Construct the habit of adapting and hone in your ability to repeatedly re-skill on greater than anyone technical skill (the very best tutor on the planet is now available to anyone, 24/7, for a low price)
- Stay near the business, not only the info
The closer you’re to the issue, the harder you’re to interchange.- Sit in additional stakeholder conversations, understand goals and constraints.
- Context will make your evaluation sharper than anything AI can infer.
- Don’t feel weird about using AI
You’re not “cheating” for those who are using a tool that makes your work higher. We’ve at all times used tools to increase human capability. This one just happens to be exponential.
Final Thought
AI will not be just one other tool in our workflow anymore.
In some ways, it’s becoming the . I imagine that while we may now not be the primary analyst on the issue, we, humans, are still those liable for asking the correct questions, making sense of the answers, and deciding what to do next. And that part still matters greater than ever.
…………
That’s it from my end on this blog post. Thanks for reading! I hope you found it an interesting read.
