A Family Tree of Software Job Titles: Comments

Former Head of DevOps Ian Miell launched an open-source visual lineage tracing eighty years of technical roles from the 1940s to 2025, according to zwischenzugs.com's report. Built with assistance from Claude, the interactive catalog maps 104 software industry titles across seven domains. The project records historical shifts from early human computers to artificial intelligence specialists.

This technical overview details the structure, mechanics, and recurrent labor patterns documented across the full software taxonomy. Engineering managers, system architects, and technical recruiters can examine how functional roles continually fragment, merge, and rebrand over five-year cycles. The historical data clarifies why modern platform disputes reflect decades of unresolved organizational divisions.

01Seven visual families and five year tiers

The interactive diagram organizes 104 software titles chronologically from top to bottom in five-year tiers, beginning with the human computer and ending with the AI agent and context engineer. Seven color-coded columns categorize roles into development, analysis and leadership, operations, security, data and AI, testing, and support. Landmark industry designations appear in bold text to highlight pivotal transitions.

Relationship paths between titles rely on two explicit connection types across the visual map. Solid arrows signify that a role directly evolved out of an established parent discipline, while dashed arrows represent cross-disciplinary influence or common career progressions. Users can isolate individual technical families or select specific titles to inspect dedicated background records and connected lineages.

The system combines a single Graphviz DOT script with a single JSON metadata catalog containing descriptions for every entry. Rendering the visual hierarchy requires executing the DOT engine to produce an SVG vector graphic. Incorporating a new title requires three edits: adding the family node, assigning the five-year row interval, and defining parent linkage arrows.

02Mathematical planning and manual machine coding

Early software history positioned machine programming as clerical assistance rather than technical design. In 1947, Herman Goldstine and John von Neumann published Planning and Coding of Problems, splitting computer operations into two responsibilities. Mathematicians were assigned the high-status task of planning problems, while coders were assigned the subordinate task of translating plans into machine instructions.

This translation work was viewed by management as lower-status clerical labor and was predominantly staffed by women. Over subsequent decades, the practical execution of writing instructions transformed into the most demanding and creative phase of technical work. Consequently, the original clerical coder designation gradually merged into the recognized programmer role.

This dynamic extended backwards into the initial hardware developments of the Second World War era. Several of the first ENIAC programmers were recruited directly from the United States Army computing corps, where human computers manually performed calculations. These workers gave their job name to the machine before transferring their skills directly into modern software programming.

03Webmaster division and full stack reunion

The webmaster role appeared across commercial computing around 1993 as an all-encompassing individual position. A single operator maintained entire websites by writing raw HTML, operating physical servers beneath office desks, answering incoming email, and troubleshooting connectivity issues. Keeping early servers running frequently required physically kicking hardware or replugging disconnected power cables.

By the mid-2000s, the webmaster designation largely vanished as organizations retired outdated terminology. Monolithic responsibilities divided across specialized groups, sending code maintenance to web developers, server hardware tasks to system administrators, and site content to digital editors and designers. This separation reflected a broader structural shift toward functional specialization.

Specialization continued around 2005 when web development split into dedicated front-end and back-end responsibilities. That division proved temporary, as industry demand shifted toward consolidating fragmented roles back together under the full-stack developer banner around 2008. This ongoing movement between decentralization and centralization mirrors recurring structural patterns observed throughout software industry history.

04Transient titles and recycled corporate classifications

Certain volatile specializations demonstrate rapid lifecycle compression across the visual timeline. The prompt engineer title gained prominence during 2023 alongside highly publicized salary listings, but advancing model capabilities quickly dispersed those skills into adjacent technical roles. Within the genealogical chart, the prompt engineer designation is absorbed into the broader AI engineer category shortly after emerging.

A similar consolidation affected scrum masters, whose hiring peaked during the 2010s before corporate structures folded duties back into delivery managers or engineering management positions. The newest role recorded on the tree is the AI agent and context engineer from approximately 2025. Historical trends suggest specialized context roles may similarly fold into broader engineering responsibilities.

Corporate environments regularly reuse historic terminology for entirely unrelated technical positions. At IBM during the 1960s, systems engineers operated as pre-sales professionals who designed customer hardware installations. During the 1990s, Microsoft MCSE certification repurposed systems engineer to identify senior server administrators, while configuration management migrated from 1960s defense hardware tracking to software version control and automation tools.

05Forward deployment and persistent operations debates

Modern enterprise expansion has revived embedded client engineering patterns under new corporate branding. Major artificial intelligence providers including OpenAI, Anthropic, and various startups are actively recruiting forward deployed engineers to integrate models inside customer infrastructures throughout 2025. This staffing model mirrors Palantir embedding engineers with customers since 2010, which itself closely resembles pre-sales installation work performed by IBM systems engineers in the 1960s.

Operational boundaries between software writing and production system reliability remain unresolved despite decades of title revisions. Google established site reliability engineering in 2003, with Ben Treynor Sloss defining the role as what happens when a software engineer designs an operations team. DevOps emerged around 2009 as a shared collaborative philosophy, yet industry hiring markets transformed DevOps engineer into an automation title by 2011.

The visual chart reflects this ongoing friction by using bidirectional arrows between site reliability engineering and DevOps designations. Platform engineering entered industry vocabulary around 2022 as a purported successor to DevOps, claiming parenthood from both disciplines while attempting to resolve persistent integration issues. Despite regular terminology rebranding, the structural division between feature developers and infrastructure operators continues without permanent organizational resolution.