Recommending matches in a text-rich, dynamic two-sided marketplace presents unique challenges due to evolving content and interaction graphs. We introduce GraphMatch, a new large-scale recommendation framework that fuses pre-trained language models with graph neural networks to overcome these challenges. Unlike prior approaches centered on standalone models, GraphMatch is a comprehensive recipe built on powerful text encoders and GNNs working in tandem. It employs adversarial negative sampling alongside point-in-time subgraph training to learn representations that capture both the fine-grained semantics of evolving text and the time-sensitive structure of the graph. We evaluated extensively on interaction data from Upwork, a leading labor marketplace, at large scale, and discuss our approach towards low-latency inference suitable for real-time use. In our experiments, GraphMatch outperforms language-only and graph-only baselines on matching tasks while being efficient at runtime. These results demonstrate that unifying language and graph representations yields a highly effective solution to text-rich, dynamic two-sided recommendations, bridging the gap between powerful pretrained LMs and large-scale graphs in practice.

Large language models (LLMs) are increasingly being deployed in high-stakes applications like hiring, yet their potential for unfair decision-making and outcomes remains understudied, particularly in generative settings. In this work, we examine the fairness of LLM-based hiring systems through two real-world tasks: resume summarization and retrieval. By constructing a synthetic resume dataset and curating job postings, we investigate whether model behavior differs across demographic groups and is sensitive to demographic perturbations. Our findings reveal that race-based differences appear in approximately 10% of generated summaries, while gender-based differences occur in only 1%. In the retrieval setting, all evaluated models display non-uniform selection patterns across demographic groups and exhibit high sensitivity to both gender and race-based perturbations. Surprisingly, retrieval models demonstrate comparable sensitivity to non-demographic changes, suggesting that fairness issues may stem, in part, from general brittleness issues. Overall, our results indicate that LLM-based hiring systems, especially at the retrieval stage, can exhibit notable biases that lead to discriminatory outcomes in real-world contexts.

Artificial intelligence is taking over countless industries around the U.S., raising concerns among Americans who fear they will be replaced by the tech. Now, new research has revealed the most and least AI-proof cities across the nation, based on five key metrics including job availability, the state's population growth rate, and job diversity. Workers based in major tech hubs should look to large, coastal metropolitan areas if they want to avoid losing out to artificial intelligence, with Phoenix, Arizona coming in first as the most AI-proof city in the country. The report warned that Providence, Rhode Island is the top city most susceptible to AI-related job loss. A report revealed the best cities to move to if you want to avoid AI and the top cities you should consider moving away from.

A reliable resume-job matching system helps a company find suitable candidates from a pool of resumes, and helps a job seeker find relevant jobs from a list of job posts. However, since job seekers apply only to a few jobs, interaction records in resume-job datasets are sparse. Different from many prior work that use complex modeling techniques, we tackle this sparsity problem using data augmentations and a simple contrastive learning approach. ConFit first creates an augmented resume-job dataset by paraphrasing specific sections in a resume or a job post. Then, ConFit uses contrastive learning to further increase training samples from $B$ pairs per batch to $O(B^2)$ per batch. We evaluate ConFit on two real-world datasets and find it outperforms prior methods (including BM25 and OpenAI text-ada-002) by up to 19% and 31% absolute in nDCG@10 for ranking jobs and ranking resumes, respectively.

Much like the internet did in the 1990s, AI is going to change the very definition of work. While change can be scary, if the last three years taught us anything, it can also be an opportunity to reinvent how we do things. I believe the best way to manage the changes ahead for employees and employers alike is to adopt a skills-first mindset. For employees, this means thinking about your job as a collection of tasks instead of a job title, with the understanding that those tasks will change regularly as AI advances. By breaking down your job into tasks that AI can fully take on, tasks for which AI can improve your efficiency, and tasks that require your unique skills, you can identify the skills you should actually be investing in to stay competitive in the job you have. This story is from the WIRED World in 2024, our annual trends briefing.

Substantial scholarship has estimated the susceptibility of jobs to automation, but little has examined how job contents evolve in the information age as new technologies substitute for tasks, shifting required skills rather than eliminating entire jobs. Here we explore patterns and consequences of changes in occupational skill and characterize occupations and workers subject to the greatest re-skilling pressure. Recent work found that changing skill requirements are greatest for STEM occupations. Nevertheless, analyzing 167 million online job posts covering 727 occupations over the last decade, we find that re-skilling pressure is greatest for low-skilled occupations when accounting for distance between skills. We further investigate the differences in skill change across employer and market size, as well as social demographic groups, and find that these differences tend to widen the economic divide. Jobs from large employers and markets experienced less change relative to small employers and markets, and non-white workers in low-skilled jobs are most demographically vulnerable. We conclude by showcasing our model's potential to precisely chart job evolution towards machine-interface integration using skill embedding spaces.

The IT industry is one of the most rapidly growing industries in the world. By 2026, its market volume is expected to reach a sensational $1.5 trillion. At the same time, Artificial Intelligence (AI) is gaining momentum as well. This innovative technology was expected to make $22.6 billion in 2020, according to Statista. Therefore, it seems that both of these industries are very attractive for tech masterminds. But the question is, how to get there?

There are two functions in this application. The first is skill identification, it can tell which parts of an input string is a skill. The second is skill classification, telling what skill it is. Essentially, they are two classification problems. One is two classes(skill/not skill) with an imbalanced dataset.

In a recruitment industry, selecting a best CV from a particular job post within a pile of thousand CV's is quite challenging. Finding a perfect candidate for an organization who can be fit to work within organizational culture is a difficult task. In order to help the recruiters to fill these gaps we leverage the help of AI. We propose a methodology to solve these problems by matching the skill graph generated from CV and Job Post. In this report our approach is to perform the business understanding in order to justify why such problems arise and how we intend to solve these problems using natural language processing and machine learning techniques. We limit our project only to solve the problem in the domain of the computer science industry.

Linking job seekers with relevant jobs requires matching based on not only skills, but also personality types. Although the Holland Code also known as RIASEC has frequently been used to group people by their suitability for six different categories of occupations, the RIASEC category labels of individual jobs are often not found in job posts. This is attributed to significant manual efforts required for assigning job posts with RIASEC labels. To cope with assigning massive number of jobs with RIASEC labels, we propose JPLink, a machine learning approach using the text content in job titles and job descriptions. JPLink exploits domain knowledge available in an occupation-specific knowledge base known as O*NET to improve feature representation of job posts. To incorporate relative ranking of RIASEC labels of each job, JPLink proposes a listwise loss function inspired by learning to rank. Both our quantitative and qualitative evaluations show that JPLink outperforms conventional baselines. We conduct an error analysis on JPLink's predictions to show that it can uncover label errors in existing job posts.