What platforms and hire-train-deploy firms hire for
Company Tier 2, Platforms and hire-train-deploy: doors, not destinations, in the staffing-and-platforms collection.
The short answer
These firms post jobs that they will fill for someone else. Talent marketplaces, like hackajob and Uplers, post the engineers they want in their pools. Freelance platforms, like Talentgigs and Instawork, post the work their clients bring. AI-data-work marketplaces, like Alignerr and Mindrift, post the experts AI companies need. So the postings show what each platform is stocking, more than what any one employer wants, and they rarely name the client.
Three platforms, hackajob, Uplers and Turing, post most of the tier's openings, and they hire nothing alike. hackajob posts Java, cloud and generalist software engineers for employers in Bengaluru, Pune and Hyderabad. Uplers posts test engineers above all, with Node, mobile and GenAI work, mostly from Delhi NCR, and it takes more juniors than anyone else here. Turing posts GenAI engineers almost entirely, with ML and Rust and C/C++ engineers beside them, from Bengaluru. Across the tier, GenAI engineers are several times their weight at product and services companies, and Java, cloud, .NET and data work are well below it. Hiring here rose and fell with those three firms, and it fell steeply after June 2026.
(More on how these firms work in Company Tier 2, Platforms and hire-train-deploy, in the staffing-and-platforms collection.)
hackajob
hackajob is a UK talent marketplace where employers approach engineers in its pool, so its postings are the roles its employers most want filled. They are mainstream backend and platform work. Java backend and full-stack engineers come first, in Spring Boot with React and Angular, Spring Security, Kafka and Kubernetes. Cloud and DevOps engineers come second, well above their usual weight, in Python and Java with AWS, Kubernetes, Terraform and Go. Generalist software engineers, whose postings name no particular stack, make up much of the rest, with Java, Python and SQL the common ground. Mobile, test, data and GenAI engineers follow in smaller numbers.
hackajob's openings are mid and senior level, with a few juniors in cloud and mobile roles. It is the platform behind Pune and Hyderabad here, as well as Bengaluru. Its posting grew from a trickle in October to December 2025 to its peak in April to June 2026, and dropped sharply in July to September 2026, when Java and cloud were most of what it posted. Few of its jobs stay open long.
Uplers
Uplers is an Indian platform that places remote engineers with overseas companies, and its postings lean towards the work those companies most often send abroad. Test engineers are by far its largest role, several times their usual weight. The postings require Java with Playwright and Selenium, GitLab CI/CD, JMeter for performance testing, REST Assured, Postman and SoapUI for testing APIs. JavaScript and Node engineers come next, in Node.js with MongoDB, Kafka and Kubernetes, then Java and GenAI engineers, mobile engineers for iOS and Android, frontend engineers in React, and Go engineers. GenAI postings at Uplers require Python with LangChain agents, LangGraph and Guardrails AI, a tool for keeping an AI's answers safe.
Uplers is the door for juniors in this tier. Its test roles take more entry and junior engineers than any other role at any firm here. It is the platform behind Delhi NCR, Mumbai and Kolkata. Uplers keeps its jobs open longer than any other firm here, with test, Node and mobile jobs posted again and again. Its posting peaked in January to March 2026, fell in April to June, and almost stopped in July to September 2026.
Turing
Turing places remote developers with companies and also supplies people for AI training work, and its postings are dominated by AI. GenAI engineers are most of what it posts, far above their weight anywhere else. Most are for software engineers who bring AI into software rather than build AI applications from scratch, and the postings require Python with JavaScript, TypeScript, React, Angular, Vue.js, NestJS and Node.js, Docker and GitHub Actions, a spread of languages that fits writing and checking code for AI models. ML engineers, also many times their usual weight, require Python with reinforcement learning and supervised learning. Rust and C/C++ systems engineers who know compilers, and Go engineers, make up much of the rest. The postings do not say which jobs build a client's product and which train an AI company's model.
Turing hires at mid and senior level, with few juniors, and nearly everything it posts is in Bengaluru. Its posting was at its height in October to December 2025, fell in January to March 2026, and all but stopped after March 2026. Its jobs are rarely open for long.
The rest of the Company Tier
The other firms post far less, and they are a mixed group. Weekday AI, a start-up that finds candidates for employers, posts a little of everything, GenAI, test, Java, Python and frontend work, mostly in Bengaluru, with a real junior intake. Crossover, which hires remote workers for a group of software companies, posts GenAI engineers who build with MCP, the Anthropic Claude and OpenAI APIs and Mistral, generalist roles that ask for process knowledge, and junior test engineers. Smart Working posts data engineers in Python, AWS Glue and Airflow, with Node and backend work. Talentgigs, a freelance and contract platform, posts Java, full-stack and ServiceNow work, much of it in Chennai, nearly all in July to September 2026. GoodSpace AI posts Node engineers, ClanX .NET engineers, Instawork Python engineers, and Talent500 hires for global companies' engineering centres in India. The AI-data-work marketplaces, Alignerr and Mindrift most of all, post GenAI experts in small numbers. The hire-train-deploy firms and training platforms, such as AccioJob, Studyecart and Revature, post very little here, though AccioJob's few postings are all for entry-level engineers. Taken together, these firms lean even further towards GenAI, Node, Python and full-stack work than the big three, and their posting was steadier through the record.
Who gets hired, by role
GenAI
Java backend and full-stack
Test
Generalists
Cloud and DevOps
JavaScript and Node
Mobile
ML
C/C++ and Rust
Also present
GenAI engineers. Much the largest role, several times its usual weight, mostly at Turing, with Uplers, Crossover, hackajob and Weekday AI behind it.
Java backend and full-stack engineers. The second-largest role but well below its usual weight, mostly at hackajob, then Uplers.
Test engineers. A little above their usual weight, mostly at Uplers, and the role that takes the most juniors.
Generalist software engineers. A large group here, mostly at hackajob and Crossover, in Java, Python and JavaScript.
Cloud and DevOps engineers. Below their usual weight, mostly at hackajob.
JavaScript and Node engineers. About twice their usual weight, at Uplers and GoodSpace AI.
Mobile engineers. Above their usual weight, at Uplers and hackajob, in Swift, SwiftUI, Kotlin and Jetpack Compose.
ML engineers. Well above their usual weight, nearly all at Turing.
C/C++ and Rust systems engineers. Mostly at Turing, in compilers and systems code.
Also present. Frontend engineers at about their usual weight. Python backend, full-stack and Go engineers. .NET engineers, far below their usual weight. Data engineers, below it. Business-platform engineers in Salesforce Sales, Service and Marketing Cloud, Oracle Integration Cloud and .NET, at Uplers and hackajob. Embedded engineers, far below their usual weight. A few ServiceNow developers and fintech software engineers.
Rare here. SAP work, security software engineers and chip designers barely appear.
The skills that set this Company Tier apart
Code for AIPython with JavaScript, TypeScript, React, Angular, Vue.js and NestJS, in Turing's GenAI postings.
AgentsLangChain agents, LangGraph, MCP, Guardrails AI and the Anthropic Claude, OpenAI and Mistral APIs, in the GenAI postings at Uplers and Crossover.
Model trainingReinforcement learning and supervised learning, in the ML postings.
Rust and compilersRust with C/C++ and compilers, in the systems postings at Turing.
Test automationPlaywright, Selenium, JMeter, REST Assured and SoapUI, in the test postings at Uplers.
Native mobileSwiftUI, UIKit, Combine, Jetpack Compose and Kotlin coroutines, in the mobile postings.
What is changing
These readings follow the record quarter by quarter, from October to December 2025, the third quarter of the financial year 2025-26, to July to September 2026, the second quarter of 2026-27. A longer record will show whether they last.
Hiring rose, then fell hardPosting here climbed to a peak in January to March 2026 and then dropped, and between April to June and July to September 2026 it fell far more steeply than at product companies, which roughly halved. The fall is Turing all but stopping after March 2026 and Uplers almost stopping in July to September. The other firms posted at a steadier pace.
The AI wave came early and recededAI work was most of what this tier posted in October to December 2025, through Turing, many times the share at product and services companies. Its share fell through 2026 as Turing stopped posting, though it stayed above the market in every quarter.
Java and cloud grewJava rose from a small role to the largest one by July to September 2026, and cloud rose with it, as hackajob became the main poster.
Testing peaked with Uplersin January to March 2026 and shrank after it.
Outside the big threeGenAI held a large place in every quarter, and full-stack and .NET work grew in July to September 2026.
Persistently open roles
Fewer jobs are persistently open here than at product and services companies, but much of that comes down to one firm.
Uplers keeps the majority of its jobs open again and again, and its test, Node and mobile jobs most of all. Native mobile skills such as Kotlin coroutines, Jetpack Compose and SwiftUI, and test tools such as JMeter and SoapUI, are the ones most often found in persistently open jobs here.
Turing's systems and ML jobs are almost never posted twice, and the same goes for the embedded and data jobs across the tier.
The other firms rarely keep jobs open, Crossover apart.
Junior jobs stay open more than any others.
How long they stay open. When a job is posted again, its first and last postings are typically about two weeks apart, and rarely more than four weeks.
The door
The door is a little wider than at product and services companies. Entry and junior hiring is above their average, and staff-level hiring is far below it. Most of what these firms post is mid level, because a platform places people who can work alone on a client's team.
Test roles take the most juniors, mostly at Uplers, with Crossover's junior test roles beside them.
Junior GenAI openings appear at Turing and Uplers, and junior cloud and mobile openings at hackajob.
Weekday AI and AccioJob take juniors across many roles.
Turing's systems, ML and Go roles, and data roles everywhere, are hired with experience.
In one line
If you test software, build in Node, write native mobile apps or work with AI in Python, these platforms are a door to remote work for clients abroad, but a door whose size rises and falls with a few firms.
Terms used on this page
Talent marketplace A platform that keeps a pool of tested engineers and matches them with companies that need them, often abroad, for a margin or a fee.
Freelance or gig platform A platform where a company hires a person for a task, a number of hours or a project, as an independent contractor rather than an employee.
AI-data-work marketplace A platform that finds experts to write examples, judge answers and correct mistakes for AI companies training their models, paid by the task or the hour.
AI training work Writing and checking the examples, code and answers that AI models learn from.
Hire-train-deploy A firm that recruits people, often fresh graduates, trains them in one technology and places them with a client while keeping them on its own payroll.
ServiceNow A platform for running a company's IT service desk and other internal requests.
MCP Model Context Protocol, a standard way for an AI model to use outside tools and data.
Reinforcement learning Training a model by rewarding good results, which is how AI models are taught to give better answers.