Data Engineering at IBM

The employer bargain at IBM is stability traded against momentum. Glassdoor sits at 3.9, a little above the middle of the pack, and the happiness tier is neutral with sentiment roughly flat, which is consistent with a company that has managed decades of restructuring without collapse but has not inspired its workforce either. Blind sentiment is mixed, which tracks: engineers who value predictability and deep client work report reasonable experiences; engineers who want fast feedback loops or aggressive career ladders get frustrated. Pay is below other Technology companies, and the ladder confirms it: $150K at L4 is where most of the 133 data points cluster. The company invests in benefits and job security relative to the market, and that trade is real, but engineers who joined expecting FAANG-style growth culture tend to say it plainly on Blind.

IBM has no tracked layoffs in the past 12 months, and low layoff risk over the next 30 days, which is a meaningful signal for a company of this size given the broader industry environment. The 2 executive departures in the past year bear watching: when IBM reorganizes a business unit, data engineering headcount tends to move with it, sometimes to client-facing roles, sometimes out. Hiring is active, with 18 open data engineering roles across 13 cities, and Austin leads the count, reflecting IBM's growing US consulting delivery footprint. The direction is cautious expansion in cloud and AI services, particularly around watsonx and hybrid cloud data infrastructure. Engineers joining now will likely be building pipelines that feed IBM's AI product layer, which means the work will change shape over the next 12 months as those products mature or stall.

IBM's data engineering work sits inside one of the most heterogeneous enterprise client bases in the industry, which shapes everything about the job. The company sells consulting, managed services, and platform tooling to banks, insurers, governments, and manufacturers, so data engineers here often own pipelines that cross client environments rather than a single internal warehouse. The visible stack reflects that breadth: AWS, Azure and BigQuery show up across listings because IBM's clients run multi-cloud and IBM's own products (Watson, watsonx) layer on top. PySpark, Python and SQL anchor the day-to-day. The work skews toward integration, governance, and reliability over greenfield architecture, because enterprise SLAs and regulatory constraints come with the territory. If you want to own a clean internal lakehouse, this is probably the wrong place; if you want exposure to a wide range of data problems across industries, it earns that reputation.