Software Job Postings Up 15%, Mostly Senior

Data Engineer View

Indeed Hiring Lab data shows software postings up 15% since February 2025, with 71% of the gain in senior roles. What it means for Data Engineer job seekers.

Published: Proudly published by: Jeff Wahl10 min read

What this post covers

01

AI-Titled Roles Account for 37% of Growth: Postings mentioning AI in the title and what they require

02

What Indeed Measured and Found: Posting index, 15% rise, and 7% overall decline

03

The Index Level: 74.4 Versus 61.1 and the Baseline: August 2026 snapshot index against May 2025 low

04

Reading Software Postings as a Data Engineer Proxy: What the adjacent category can and cannot say

05

Why Indeed Says Causation Is Not Shown: Author's caveat on attributing the rebound to AI

06

Seniors Took 71% of the Gain: Seniority split of the May 2025 to May 2026 increase

07

What the Senior Tilt Means for Interview Preparation: Early-career positioning and senior-level loop expectations

US software development job postings on Indeed are up almost 15% since February 2025, while overall postings fell about 7% over the same stretch, according to Indeed Hiring Lab economist Guillermo Gallacher.[1] For anyone tracking the data engineer job market 2026, the more important number is a different one: senior roles accounted for 71% of the net increase in software development postings from May 2025 to May 2026.[1] Postings also remain roughly 27.5% below their pre-pandemic level.[1]

So software hiring is recovering, and most of that recovery is going to experienced engineers. Indeed doesn't track Data Engineer as its own category, which makes this an adjacent signal. It's still a useful one if you're early-career or mid-level and trying to figure out what your next loop will look like.

A note on sourcing. The numbers here come from Indeed Hiring Lab's July 2026 publications and the Indeed series published on FRED. CNBC's October 2026 coverage reproduces the same headline figures and credits Hiring Lab.[3] Some secondary write-ups use different dates and baselines for these numbers. I've stuck to Indeed's figures and flagged the places where the time windows differ.

Prepare for the interview
01 / Open invite
02min.

Know the patterns before the interviewer asks them.

a system design query, the same shape a screen would give you.
The diff against expected. Where ties broke. What you missed.
sandbox
1source → bronze → silver → gold
2 ingest : CDC + Kafka
3 transform : dbt + Airflow
4 serve : Snowflake
5
Execute your solution0.4s avg.
PayPalInterview question
Solve a problem

What Indeed Hiring Lab Measured on Software Development Job Postings

Indeed's posting index counts seasonally adjusted job postings relative to February 1, 2020, which is set to 100. The software development series is public on FRED under the ID IHLIDXUSTPSOFTDEVE.[6] When I say "postings are 27.5% below baseline," I mean the index sits around 72.5.

Gallacher's July 8, 2026 analysis uses February 24, 2025 as its anchor date. Software development postings have risen almost 15% since then, while overall postings fell 7%.[1] The comparison to before the pandemic looks different. Overall postings are roughly back to their February 2020 level, and software is still about 27.5% short of it.[1]

That is a recovery. It also still leaves software well below where it was in 2020. Both of those statements are true.

How the gap has moved

As ofMeasureValueSource
Jan 23, 2026Software dev postings vs Feb 202031.8% belowHiring Lab Tech Q4 report[4]
Jan 23, 2026Software dev postings, year over year+0.8%Hiring Lab Tech Q4 report[4]
Mid-2026Software dev postings vs Feb 202027.5% belowHiring Lab, July 8[1]
May 1, 2026FRED index level72.14FRED[6]
Sep 11, 2026FRED index level76.62FRED[6]
Oct 23, 2026FRED index level80.79FRED[6]

Moving from 31.8% below baseline to 27.5% below is consistent with a recovery; the 2 figures just come from different dates. If the most recent FRED reading of 80.79 is accurate, the gap has narrowed to roughly 19% since Indeed's July write-up. I'd pull the live table before relying on that, because indexes like this get revised.

You can check this yourself. Load the FRED series into any warehouse and run something like the query below. It also happens to be decent interview practice, since it combines a baseline comparison with a window function.

-- Assumes the FRED series IHLIDXUSTPSOFTDEVE is loaded as
-- indeed_softdev(obs_date DATE, idx NUMERIC), Feb 1 2020 = 100
WITH anchored AS (
SELECT
obs_date,
idx,
FIRST_VALUE(idx) OVER (
ORDER BY obs_date
) AS idx_at_anchor
FROM indeed_softdev
WHERE obs_date >= DATE '2025-02-24'
)
SELECT
obs_date,
idx,
ROUND(idx - 100, 1) AS pct_vs_feb_2020,
ROUND(
100.0 * (
idx - idx_at_anchor
) / idx_at_anchor,
1
) AS pct_since_anchor
FROM anchored
ORDER BY obs_date DESC
LIMIT 10

If FIRST_VALUE with an ORDER BY window isn't automatic for you yet, spend an afternoon on window function practice. That pattern appears in a lot of SQL rounds.

Seniors Took 71% of the Software Development Job Postings Gain

This finding is the one candidates should pay attention to. Of the net increase in software development postings from May 2025 to May 2026, 71% came from senior roles. Another 37% came from postings with AI in the job title.[1] The 2 groups overlap, because a "Senior AI Engineer" posting counts in both. Don't add them together.

Pay attention to the time windows as well. The 15% rebound runs from February 2025. The 71% senior share covers May 2025 to May 2026. Those are separate periods, and a sentence that blends them is making a claim neither number supports.

Indeed's companion piece on seniority shows the same pattern in the overall mix of postings. In Q1 2026, senior-level roles made up 69.3% of software development postings, and entry-level roles made up 4.5%.[2] The same article reports that, as of May 2026, senior postings were up 14.7% year over year and entry-level postings were down 7.5%. It also says mid-level postings were 6.7% below January 2025.[2]

Indeed assigns seniority with an algorithmic text classifier. Each posting gets exactly one label (Entry, Mid, or Senior), and Hiring Lab recently changed its method to classify based on the responsibilities and training a job requires.[2] The change matters in 2 ways:

  • Comparability: Older Hiring Lab seniority analyses used a different method, so don't line them up against the 2026 numbers as if they measured the same thing.
  • Titles vs. text: The label is based on what the posting asks for. A role titled "Data Engineer II" can be classified Senior if the description reads like senior work. Recruiters read postings that way too, in my experience.

I've watched this happen from the hiring side. In one cycle, our team had 3 senior reqs open and zero junior ones. The reason wasn't distrust of junior engineers; nobody had the time to onboard them. When a team is under pressure to ship, it tends to hire someone who already knows how to deal with a failing backfill at 2am.

Reading Software Postings as a Proxy for Data Engineer Hiring Trends

Indeed has no Data Engineer category, so any claim about data engineer hiring trends drawn from this data involves an inference. The closest category Indeed tracks is Data & Analytics, and it looks weaker than software. As of January 23, 2026, Data & Analytics postings were 10.1% lower year over year and 40.1% below February 2020, according to Hiring Lab's Q4 tech report.[4] The table in that PDF is dense, so check the column labels before you quote it.

Data & Analytics is also where AI terms show up most. Nearly 45% of Data & Analytics postings contained AI-related terms as of December 2025, compared with about 15% in marketing and 9% in HR.[5]

Here's how I read it. DE roles fall across both of Indeed's categories. Some companies post a Data Engineer role that is essentially backend software engineering with Spark. Others post the same title for SQL plus a dbt project. The title means different things depending on who's hiring. That ambiguity is a feature of the industry, and it's also why neither Indeed series measures DE demand cleanly.

Keep 3 limits in mind:

  • Postings measure listings. Indeed's index counts openings that employers advertise. It doesn't count offers or hires, so a 15% rise in postings doesn't mean 15% more offers.
  • The category boundary is fuzzy. A DE posting could fall into software development, into Data & Analytics, or be split across both, depending on how it's written.
  • The level is still low. Software postings remain below 2020, and the closest data category was further below baseline as of January.[4] Plan for a competitive market, even with volume improving.

Indeed Says Causation Isn't Shown, and You Shouldn't Claim It Either

Gallacher's piece chose February 24, 2025 as its anchor, and secondary coverage ties that date to the launch of Claude Code. Indeed's own wording is careful, though. The article states that correlation does not imply causation, while reporting that the measured relationship is statistically significant.[1]

Hiring Lab's seniority article also brings up a competing explanation. It notes that New York Fed researchers argued remote work is more closely tied to recent graduates' job struggles than AI is.[2] I haven't read the underlying Fed paper, so I'm passing that along as Hiring Lab described it.

My view: I've lived through 3 waves of "data engineering is getting automated away," and I'm still employed and still debugging the same kinds of problems. Each wave got a tidy causal story from people who wanted one. The honest description of what we know is this: software postings bottomed out, began rising around the time agentic coding tools arrived, and the increase has been concentrated in senior roles and roles with AI in the title. That's an observed pattern. The causal explanation isn't settled.

This affects interviews too. If a hiring manager asks how AI is changing the job, avoid answering with a confident causal claim based on a postings chart. Describe what you've actually seen in your own work. Coming from someone they might hire, that answer carries more weight.

Analysts Are Slowing the Store Down

> Analytics queries against the production database are slowing the live application. Move analytics onto its own warehouse fed from the database's change log, while a merchant dashboard shows new orders within fifteen minutes on a path of its own.

Source
Transform
Storage
Quality
Consumer
Queue
Zone
Pipeline Architecture
Sketch the architecture

Add a node from the toolbar or right-click the canvas

Drag from a right port to a left port to wire data flow

Entry Level Data Engineer Jobs: How to Position Yourself

If you're early in your career and your applications keep disappearing without a reply, the data supports your experience. An entry-level share of 4.5% of software postings is very small.[2] The difficulty you're running into is real.

I've long believed Data Engineering isn't a true entry-level job anyway. It combines business context, analytics, infrastructure, software engineering, and on-call work. Most people I know got into DE from an adjacent role, and that route still works. If you're coming from analytics, the analyst to Data Engineer path is well documented.

What I'd do with the current numbers:

  • Write your resume in terms of responsibilities. Indeed's classifier, and the people screening applications, look at what you owned. Don't write "familiar with Airflow." Write "owned 14 DAGs, on-call for the ingestion layer, cut the daily failure rate in half." Our Data Engineer resume guide covers this in detail.
  • Show work that looks like production. Tests, retries, alerting, and a backfill strategy. A side project that can be rerun safely tells me more than a dozen tool logos. If you haven't built an idempotent pipeline, start there.
  • Target postings for data roles specifically. The software category is mostly senior. Search for the actual DE titles and read the posting text to see what level they really want.

I'm in favor of everyone using AI tools; the companies will be. Still, a candidate who can't explain why their own pipeline dropped rows is going to struggle against people with 6 years of production incidents behind them.

Senior Data Engineer Demand Means Senior-Style Loops

The mid-level group has the hardest position here, more so than juniors. Mid-level postings dropped against January 2025 while senior postings increased.[2] If you have 3 or 4 years of experience, the openings that are growing are the ones written for engineers a level above you. Compete for them anyway. Just know the loop will be calibrated to that level.

I've seen people with 10 years of experience get downleveled because they couldn't explain their design decisions under pressure. Interviewing is a different skill from doing the job. With senior Data Engineer demand making up most of the growth, expect these rounds:

  • Data modeling with tradeoffs. Grain, slowly changing dimensions, and when a wide denormalized table is cheaper to run than a star schema. Work through data modeling interview questions until you can defend the grain choice before they ask you to.
  • Incident postmortems. "Tell me about a time a pipeline silently lost data." Have one with numbers ready: what dropped, how you found it, what you changed so it couldn't happen again.
  • Cost decisions. Storage is about 2 cents a GB and engineers are expensive. Senior candidates reason from that, while junior candidates tend to reason from "best practices."
  • Verifying AI-assisted code. With 37% of the software gain in AI-titled roles,[1] expect to be asked how you check SQL or pipeline code an assistant generated. "I read it carefully" won't be enough. Describe the row-count reconciliation, the null checks, and the test you added.
The rebound is real, and it's concentrated at the senior level. Whatever your title says, prepare for the loop that sits 1 level above it. Concepts like modeling, debugging, and cost reasoning are what make a mid-level candidate come across as senior; knowing an extra tool doesn't do much.

The most useful practice I've found for this is talking through decisions aloud with a timer running. A mock interview simulator forces that in a way that reading answers silently never will.

What to Watch in the Data Engineer Job Market Next

Here's what I'd keep an eye on over the next few months, all from data you can check yourself:

  • The FRED series. If the software index keeps rising toward 85 or above, the gap to 2020 is closing quickly. If it levels off around 80, the rebound was a one-time correction.
  • The entry-level share. 4.5% is the figure to compare against in Indeed's next seniority update. Any movement above that would be the first good news for juniors in a while.
  • Data & Analytics against software. If the data category starts catching up to software, the proxy gets more useful. If it keeps lagging, DE hiring is tracking the weaker of the 2 series.
  • The AI-in-title share. Watch whether it keeps growing or whether AI becomes a normal line in ordinary DE postings. With nearly half of data postings already mentioning AI terms,[5] my bet is the second.

In the meantime, prepare as if the bar is going up, because the evidence suggests it is. Practice medium-difficulty SQL and Python until they're routine; working through 50 or so practice problems covers most of what companies ask. Then spend most of your remaining time on modeling and pipeline architecture, since those rounds are where senior-style loops get decided. Our Data Engineering interview prep guide lays out the full sequence.

Tools change about every 18 months, while problems like schema drift, late-arriving data, and upstream teams breaking contracts stay the same. The market is currently rewarding engineers who've dealt with those problems before, so build that experience and be ready to describe it in interviews.

References

  1. Indeed Hiring Lab, Guillermo Gallacher, "AI and Job Postings: From Destruction to Creation?," July 8, 2026. hiringlab.indeed.com
  2. Indeed Hiring Lab, "The Labor Market Is Tilting Toward Seniority," July 23, 2026. hiringlab.indeed.com
  3. CNBC, "AI is automating coding. So why are software developer job postings on the rise?," October 11, 2026 (reporting Indeed Hiring Lab data). cnbc.com
  4. Indeed Hiring Lab, "HiringLab Tech Q4" report (PDF), February 2026. d341ezm4iqaae0.cloudfront.net
  5. Indeed Hiring Lab, "January 2026 US Labor Market Update: Jobs Mentioning AI Are Growing Amid Broader Hiring Weakness," January 22, 2026. hiringlab.indeed.com
  6. Federal Reserve Bank of St. Louis (FRED), "Software Development Job Postings on Indeed in the United States (IHLIDXUSTPSOFTDEVE)," data from Indeed Hiring Lab. fred.stlouisfed.org

data engineer job market 2026software development job postings Indeeddata engineer hiring trendssenior data engineer demandentry level data engineer jobs
02 / Why practice

Try the actual problems

  1. 01

    Reading a solution is not the same as writing one

    Every engineer who has frozen on a query they had read a dozen times knows the gap. The only preparation that closes it is producing the answer yourself, under time, before the interview does it for you

  2. 02

    76% of hiring managers reject on the coding task, not the resume

    From HackerRank's 2024 Developer Skills Report. Candidates who look strong on paper still fail the live screen if they haven't done timed, executable practice

  3. 03

    System design comes down to the calls you defend out loud

    Ingestion, batch vs streaming, the bronze/silver/gold layers, idempotency, backfill and replay. Sketching the pipeline and naming the failure modes is the signal, not the boxes