Top programming languages in 2026: data-backed ranking

This piece cross-references the TIOBE Index, the Stack Overflow Developer Survey, and job market signals to build a ranking that actually holds up.

Top programming languages in 2026: Quick ranking

Python holds the #1 spot on the TIOBE Index (18.53%, August 2026, source), the longest-running programming language popularity tracker, which measures search engine query trajectory rather than actual language use in production. TIOBE's companion PYPL index tells a similar story: Python's climb has been steady since 2020, driven by data and machine learning models, not web development.

Ranking by TIOBE Index alone produces a different order than ranking by Stack Overflow Developer Survey responses. Each dataset weighs a different signal: search interest versus what developers report actually using day to day, so the top list shifts depending on which one you trust. TypeScript is the clearest example: it sits at #44 on TIOBE (0.37%) but is used by 48.8% of professional developers per the 2025 Stack Overflow survey, one of the largest gaps between search interest and real adoption on this list.

Hiring difficulty and popularity rank also move independently. A language can sit high on every index and still be the hardest developer role on a team to fill this quarter, that's the case for Rust and Go against their job-posting volume, addressed below.

The general pattern holds across sources, with one caveat every senior engineer already suspects: JavaScript still owns front-end and full-stack web development by sheer library and framework abundance.

Go has become a common choice for distributed and operating systems infrastructure. Rust keeps climbing wherever memory safety in systems-level code outweighs a steeper learning curve.

Visual Basic is the index's biggest surprise: it still sits at #7 (2.18%), ahead of SQL, Rust, PHP, and Go, despite near-zero use in new systems work. That's TIOBE's search-volume methodology at work, not a sign VB is still a common choice for new projects, enterprises maintaining legacy desktop applications keep query volume high even as greenfield use approaches zero. Delphi (#11, 1.37%) follows a similar pattern.

The table below is a directional composite, not a strict merge of any single index's exact rank order. Treat "movement" as a general trend rather than a precise delta, since TIOBE and Stack Overflow use different methodologies and don't always agree on exact placement.

Quick win: staffing a web feature this quarter favors JavaScript or TypeScript for speed of hire. Using Rust for a performance-critical service means budgeting a longer recruiting timeline than a Go or Python hire, a pattern that holds industry-wide, not just anecdotally.

This is where community signals genuinely help. A large, active developer community, visible in forum threads, GitHub stars, and conference talks, usually correlates with easier hiring even when raw ranking numbers disagree. If you're weighing related languages for a new project, cross-check at least two independent sources before committing a roadmap, and expect a few minutes' read on each to save weeks of hiring pain later.

How we ranked these languages: TIOBE, Stack Overflow, and job data

The TIOBE Index and the Stack Overflow Developer Survey rarely agree, and most listicles quietly pick whichever one flatters their thesis. We use both, plus job market demand data, because each measures a different thing.

TIOBE Index tracks search engine query volume for a language name plus "programming": a proxy for how much developers are researching or learning it, not how much production code runs on it. That is why Visual Basic and other legacy languages still post a real share on TIOBE despite near-zero use in new systems work.

Stack Overflow's survey asks working developers what they actually use day to day, which explains why JavaScript and Python dominate there while TIOBE's basic query-trajectory model also rewards languages with heavy tutorial and course traffic.

Job market demand, postings volume and salary data, is the lagging indicator: it tells you what companies are staffing for right now, typically behind what's trending in surveyed developer sentiment. We weight multiple sources rather than rank on TIOBE alone, because TIOBE's own methodology page states plainly that it does not measure the best language or the amount of code written in it.

Practically, this fanout matters for hiring: teams recruiting for Rust or Go often find candidates take longer to source than job-posting volume alone would suggest, because demand has outpaced the trained talent pool.

Which index should you actually trust for a hiring decision?

Match the index to the decision, not the other way around. Three scenarios come up repeatedly in stack and hiring conversations:

• Scoping a multi-year platform bet (which language will still have a healthy talent pool in five years): weight TIOBE's long-run trajectory over any single year's number, it smooths out hype cycles better than a one-year survey snapshot.
• Hiring or training a team this year: weight the Stack Overflow Developer Survey's usage figures over TIOBE. It tells you what working developers actually reach for day to day, which is closer to who's available to hire.
• Estimating how hard a specific role will be to fill: weight job posting volume and time-to-fill data over either index. Neither TIOBE nor the survey tracks live hiring demand directly.

A defense or aerospace systems project ranked on search volume alone would be a mistake; those decisions live closer to the survey and job posting data than to TIOBE's query counts. TypeScript is the sharpest illustration of why this matters: #44 on TIOBE (0.37%) but used by 48.8% of professional developers per the Stack Overflow Developer Survey 2025, a team hiring off the TIOBE number alone would badly underestimate how mainstream it already is.

Python: Why it still leads every index

Python has topped the TIOBE Index since late 2021 and still leads as of August 2026, ahead of C, C++, and Java.

On the Stack Overflow Developer Survey, it's used by 54.8% of professional developers, the second-highest usage figure behind only JavaScript, a signal of sustained day-to-day adoption rather than pure legacy inertia.

The reasons haven't shifted much year over year. Python's learning curve is the shallowest of any language with comparable reach: readable syntax, dynamic typing, and a library set (NumPy, Pandas, FastAPI, PyTorch) that covers data science, web back ends, and automation scripts with the same core language.

Visual Basic still shows up a few spots below it on TIOBE, a reminder that search-volume-driven indices reward beginner tutorials as much as production share.

Software developers in the US earn a median of $132,684 per year (U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics, May 2025); Python roles specializing in data and machine learning tend to land at or above that benchmark, general-purpose Python roles track closer to the median.

Python's weak spot shows up at the concurrency and latency ceiling, a pattern that recurs across the industry: teams keep Python for data pipelines and model training, then peel off the request-serving hot path into a faster language once GIL contention or tail latency becomes the bottleneck.

JavaScript and TypeScript: The web's default and its typed superset

JavaScript remains the language of the web front end, full stop. It sits at #6 on TIOBE (2.63%, August 2026), a spot it has held with minor share swings for years. TypeScript tells the more interesting story: it ranks a lowly #44 on TIOBE (0.37%) yet is used by 48.8% of professional developers per the Stack Overflow Developer Survey, a top-four usage figure. Nobody searches "TypeScript tutorial" at the rate they search "Python tutorial," but plenty of teams are shipping TypeScript in production, TIOBE's search-proxy methodology simply can't see that.

We treat this as one language with an increasingly mandatory typed layer for anything beyond a prototype. Most new front-end and Node.js service work defaults to TypeScript; plain JavaScript survives mainly in legacy codebases, quick scripts, and teams too small to justify a build step.

Teams moving off legacy PHP or Joomla platforms increasingly pick a typed JavaScript stack over plain JavaScript specifically for the compile-time guarantees on data shapes crossing API boundaries, not for syntax preference, see the Anime Digital Network case below for a real example of that kind of move.

Job postings that name TypeScript explicitly tend to pay a premium over generic "JavaScript developer" listings. The learning curve is a real cost, though: engineers who never used a statically typed language before JavaScript often need a few weeks to stop fighting the compiler rather than working with it.

For teams building anything customer-facing on the web in 2026, TypeScript is the default we recommend, with plain JavaScript reserved for narrow, throwaway use cases. If the hiring math points to TypeScript but internal expertise is thin, many teams choose to hire a TypeScript team rather than wait out the learning curve.

C, C++ and Java: Enterprise and systems programming holding position

C holds #2 on TIOBE (11.10%, August 2026), the highest of the three, because firmware, embedded controllers, kernel-level code, and automotive/aerospace systems, where a runtime or garbage collector can't exist, keep its search interest high even though almost no new application-layer code gets written in it.

C++ sits at #3 (8.62%) just ahead of Java. Its story diverges from C in one clear way: C++ owns application-level systems programming where performance matters but some abstraction is tolerable, game engines, trading systems, and robotics stacks are typically built with it.

Java holds #4 (8.25%), a rank it has defended for years despite Kotlin eating into Android work and Go pulling backend greenfield projects away from it. It is not growing. It is not shrinking much either, and in enterprise systems, that stability is the whole appeal.

Embedded C and C++ roles are consistently among the harder positions to fill industry-wide: fewer candidates apply per posting than for Python or TypeScript roles at equivalent seniority, and that scarcity is reflected in above-median compensation relative to the $132,684 BLS software developer benchmark.

None of the three is trending down enough to worry about, and none is trending up enough to call a resurgence. For a CTO scoping a new build, that flatness is the signal: pick Java or C++ for depth of existing codebase and library maturity, pick C only when the hardware constraint forces your hand.

C# and .NET: Current trajectory

C# sits at #5 on TIOBE (4.09%, August 2026), and its trajectory is quietly upward because it is entrenched in legacy systems, not because of new adoption. Unity keeps C# the default language for game development, Azure keeps it wired into enterprise cloud contracts, and Blazor has given it a real, if still minority, claim on web front-end work.

Teams running long-lived Visual Basic or classic ASP.NET codebases typically keep the language and replace the runtime when they modernize, moving onto current .NET rather than rewriting business logic from scratch. The migration cost then sits mostly in tooling and CI, not in touching the business logic itself.

Job market demand for C# is steady rather than hot, positioning it as a safer, if lower-paid, bet than most scripting languages, though below Rust or Go in typical per-role compensation. The Stack Overflow Developer Survey shows C# used by 29.9% of professional developers, a solid mid-tier share with less year-over-year volatility than JavaScript or Python.

The learning curve is moderate: straightforward for anyone coming from Java, harder for developers used to dynamic typing. Senior .NET architects who also know distributed systems and cloud-native patterns are the tightest part of the market, that's where most current demand concentrates.

If senior .NET expertise is scarce internally, many teams bring in staff augmentation support to cover distributed systems and cloud-native work without a lengthy hiring cycle.

Go vs Rust: Which one should you bet on for backend and systems work?

Go and Rust rarely compete for the same hire. Go wins when you need concurrent network services shipped fast by a mid-size team; Rust wins when a memory-safety bug or a millisecond of latency is unacceptable. The learning curve is the real fork in the decision, not raw performance.

Go's goroutine model and garbage collector make it approachable for developers coming from Python or Java, most engineers ship production Go within a few weeks. Speed to productivity is Go's whole pitch. Rust's borrow checker enforces memory safety at compile time, which eliminates a class of bugs entirely but typically extends onboarding to two to three months for mid-level hires.

On demand, Rust has topped the Stack Overflow Developer Survey as the most admired language for several years running, but Go shows up more often in job postings for platform and infrastructure roles. The two sit on opposite sides of TIOBE's top 10, too: Rust holds #10 (1.45%), Go sits at #14 (1.07%), a gap worth flagging since TIOBE rank and real hiring demand diverge sharply for both languages, Go's actual professional usage (17.4% per Stack Overflow) outpaces what its TIOBE position would suggest.

Migrating a request-serving hot path from Python to Rust only pays off when the latency or memory-footprint requirement is real, not aspirational, that trade shows up repeatedly in performance-driven rewrites across the industry. If your bottleneck is engineering throughput rather than runtime cost, Go's shallower learning curve gets a team productive faster.

SQL: Still non-negotiable for data-heavy roles

SQL rarely shows up in language popularity debates because it isn't a general-purpose programming language, yet job market demand for it outranks most languages on this list. It sits in a category of its own: a query language that every data-heavy application, analytics pipeline, and backend system depends on regardless of which programming language wrote the surrounding logic.

The TIOBE Index tracks SQL alongside general-purpose languages like Python and JavaScript, and it holds #8 (1.88%, August 2026), a top-10 share of search-engine mentions used as TIOBE's popularity proxy, a strange bedfellow next to Visual Basic and Go, but the comparison shows how much of real-world development still runs through relational systems.

SQL fluency is a baseline skill worth checking for regardless of the primary stack: a Rust backend, a TypeScript front end, a Python data service all still hit a Postgres or MySQL instance eventually.

Per the Stack Overflow Developer Survey, SQL is used by 61.3% of professional developers, one of the most widely used languages in the survey, a figure that has held steady across multiple survey years. The learning curve is shallow at the basic query level and steep at the systems level, where query planning and indexing separate competent developers from ones who can actually scale an application.

Swift vs Kotlin: Which wins for mobile development?

Kotlin wins on reach, Swift wins on depth, and TIOBE's ranking of the two runs counter to what usage data would suggest.

Kotlin covers Android, backend development via Ktor, and increasingly iOS through Kotlin Multiplatform, so teams standardizing on one language across mobile and server pick it more often, it's used by 11.5% of professional developers per the Stack Overflow Developer Survey 2025, roughly double Swift's 5.7%. Yet on TIOBE, Kotlin sits at #28 (0.64%), well behind Swift's #17 (0.96%), the reverse of what real-world adoption implies, another case where TIOBE's search-volume proxy diverges from actual usage.

Swift remains the stronger choice for teams building iOS-only products that need tight integration with Apple's frameworks and SwiftUI. Its higher TIOBE position likely reflects heavier tutorial and course search traffic among the large population of iOS learners, not broader production reach than Kotlin.

Both languages solve the same problem Java and Objective-C used to own less elegantly: modern concurrency and null safety by default. Swift uses structured concurrency and actors, while Kotlin uses coroutines, each helps developers avoid the callback complexity that plagued earlier mobile stacks.

Job market demand tilts toward Kotlin simply because Android's install base is larger globally, and companies reuse Kotlin engineers on backend work. Netguru used Kotlin Multiplatform and SwiftUI together to ship a single mobile codebase across iOS and Android for the Pismo Foundation, a journalism platform that needed to launch fast without maintaining two native builds, the app hit a 4.5-star App Store rating and 30% of subscribers downloaded it within the first month, with users spending nearly four times longer in-app than on the desktop site.

Kotlin Multiplatform is the right call when a team needs to ship Android and iOS logic from one codebase without duplicating business rules; pure Swift still wins when the product roadmap is iOS-first with no near-term Android plan.

Neither language is disappearing from job postings soon. The choice comes down to platform strategy, not TIOBE's rank order, which is a search-interest artifact here more than anywhere else on this list.

PHP and R: Niche but still relevant

PHP holds #13 on TIOBE (1.11%, August 2026), a rank that surprises engineers who assumed it was fading years ago; it's used by 18.9% of professional developers per the Stack Overflow Developer Survey 2025, ahead of Go and Rust on raw usage even though neither TIOBE nor daily developer sentiment treats it as a growth language.

R sits further down most indexes, yet it still commands a loyal community. Statisticians, academics, and data science teams keep using R because its libraries handle statistical modeling better than most general-purpose languages.

Neither language sees much use in new systems development. PHP's footprint today is legacy: WordPress, Laravel, and web applications built a decade ago still carry a large share of production traffic. Visual Basic and Delphi tell a similar story, both still present in the rankings, both still running back-office systems nobody wants to touch, kept alive by enterprises unwilling to fund a rewrite.

Netguru's work modernizing Anime Digital Network, a streaming platform running on legacy PHP and Joomla, is a real example of what that kind of move looks like in practice: downtime during traffic spikes and an aging stack that blocked new features forced a full rebuild onto Node.js and React, with AWS-based cloud infrastructure and microservices replacing the monolith. The result was a platform that supports zero-downtime releases and autoscaling for continued growth, the kind of outcome that's hard to bolt onto a legacy PHP monolith without a genuine rewrite.

Mobile

Swift remains the native choice for iOS, Kotlin for Android, and both keep strong TIOBE Index momentum as platform vendors invest in tooling. Cross-platform teams increasingly reach for TypeScript-based frameworks instead of maintaining two native codebases.

AI and data

Python is still the default for AI, ML, and data pipeline work, carried by PyTorch, pandas, and a ten-plus-year head start on tutorials and package maturity. Distributed data and analytics teams building on top of Python rarely have a reason to switch mid-project.

Embedded and systems

Rust is displacing C and C++ in memory-safe, performance-critical systems: automotive, aerospace, and defense codebases increasingly specify it for new modules, particularly where a request-serving or control-loop hot path needs throughput guarantees Python or C can't offer without extra tooling.

Game development

C++ still runs the engine layer through Unreal, and C# covers Unity-based studios; neither shows up as often in general software hiring data as web or backend roles.

Blockchain

Rust and Go power blockchain infrastructure (validator clients, node software), while Solidity remains the standard for smart contract logic itself, a narrower, domain-specific skill set.

DevOps

Go anchors most DevOps tooling: Docker, Kubernetes, and Terraform are all written in it, and its static binaries and fast build times make it the practical choice for infrastructure automation. Senior Go and Rust engineers are consistently among the hardest infrastructure roles to fill industry-wide, well ahead of TypeScript or Python hires.

Comparison table: Use case, learning curve, popularity rank, and real-world usage

The table below ranks the top seven programming languages on TIOBE Index position, Stack Overflow Developer Survey adoption among professional developers, and learning curve for 2026 hiring decisions.

Methodology matters here. The TIOBE Index measures search-engine query volume for a language name, a proxy for interest and tutorial traffic, not actual production usage. The Stack Overflow survey asks working developers what they use daily. The two diverge sharply for TypeScript, Go, and Kotlin in particular, each ranks far higher on real-world usage than on TIOBE's search-volume proxy. Use the TIOBE column to gauge long-run mindshare and tutorial demand, and the Professional Usage column to gauge who you can actually hire this year, the Pay Signal column then tells you what that hiring difficulty costs.

Pay tracks scarcity more than raw demand. Rust commands a premium not because job postings are numerous, but because few developers have production experience with its ownership model and borrow checker, that scarcity, not popularity, is what shows up in comp.

FAQ: Choosing a programming language to learn or hire for in 2026

Need help choosing the right stack for your project?

Picking between Rust and Go for a performance-critical service is a real engineering tradeoff, not a TIOBE Index lookup. The right call depends on whether the bottleneck is latency and memory footprint, where Rust's guarantees earn their onboarding cost, or engineering throughput, where Go's shallower learning curve gets a team shipping faster.

If your team is weighing language, stack, and development approach against a real deadline, we can give straight answers on what fits your architecture and hiring reality, not a generic index score. Talk to our team before you commit to a stack.