Every technological advancement has benefitted hugely from the communities that have been formed around it. Communities nurture developers and evangelists in a way that strengthens the domain as a whole.
The ecosystems around the world are now ripe for widespread adoption of ML as a service. Today, there are hundreds and thousands of ML communities around the world, but only a handful of communities have a positive impact.
So, what does it take to build an efficient and functional ML community?
Connect Through Conferences
There is no top limit for what conferences have to offer. The opportunities are only bound by the need of the attendees. From getting to meet experts to landing a job and closing business deals, conferences work differently for different people. Especially for a relatively new and diverse population like that of AI, it is important to provide a conference that offers 360-degree coverage of all the departments involved.
So far, conferences like NeurIPS, ICLR or Google’s flagship events have been satiating the need of the global audience. Though India still has a new scene, there are a handful of conferences that really pack a punch. One such is the Machine Learning Developers Summit (MLDS) organised by Analytics India Magazine every year in the month of January. In its latest edition, MLDS hosted the likes of Julia creator Viral Shah along with other industry insiders. MLDS is one of the four largest AI conferences conducted by AIM along with Cypher, Machinecon and Rising.
Conduct Hackathons
There can be nothing more effective than an opportunity to test the skills of those who are involved. Machine learning is still in its infancy, and the practitioners and aspirants do not have huge technical gaps, unlike in other domains. So consistent participation in hackathons can put the rookies on par with the industry experts (excluding doctorates and postdocs).
Kaggle has set an unprecedented example of how to build a community around hackathons. It single-handedly brought together people from various backgrounds to work on data-driven solutions. It is no exaggeration that Kaggle has made the young ML population comfortable with code and algorithms.
Analytics India Magazine’s Machine Hack consistently offers practitioners to test their skills by conducting a variety of hackathons. From patient drug switch prediction to food delivery time prediction to visualisations of India pollution data, Machine Hack touches upon skills ranging from text classification to sentiment analysis to time series modelling and even visualisation skills. These hackathons are tailored made to encourage the participants to come with solutions that can be showcased at any machine learning product based company. The community that Analytics India Magazine engages, exemplifies how an all-round approach benefits the community.
Conduct Niche Meetups
Conferences are a good avenue to create awareness amongst newcomers, connect cross-domain experts and other players in the industry. However, as one makes progress through theory, a good hands-on session on machine learning tools can be a great boost. So, initiatives are to be taken where the supply-demand gap is filled on both ends—industry and individual.
Analytics India Magazine frequently collaborates with machine learning giants like NVIDIA and Intel to conduct meetups that brings together a crowd that is segmented based on their enthusiasm to know more about specific tools.
AIMixer is another such event that hosts leaders who are spearheading transformation. This event features decision-makers including CIOs, CTOs, CMOs, CDOs and Chief Analytics Heads across different organisations and covers specific challenges and opportunities in analytics, AI and automation across banking, insurance and finance domains.
In this way, the industry leaders and the practitioners get to know the skill gaps. These meetups not only provide an opportunity to learn certain technology but also acts as a platform for collaboration with other interested parties.
Reward Those Who DO
Any community pivots around the pioneers and the prime movers. In the case of machine learning, it is the experts who incorporate AI principles into their product pipelines, the developers who build high-end APIs and the organisations that make their innovations public. These are the people that have to be rewarded handsomely in all decent ways possible.
This is a favour to the community itself as rewarding or recognising establishes an environment that incentivises risk-takers and innovators to go full throttle. The longevity of technology relies entirely on breakthroughs, so it is an imperative of the community to diligently recognise its own tribe.
Eliminate Zero-Sum Games
A community will eventually implode in the absence of net positive gain. Every member joins a community to gain something. It can be to satiate their curiosity or to mint money out of knowledge gained. The reasons can be plenty, but for those who lead the community, the sustainability lies in their consistency. For example, Kaggle provides a platform to both the industry and the individual to collaborate to solve hard problems. The outcome is tremendously positive!
A solution can bring huge revenue to organisations and can also open up the world of opportunities for developers. It is highly likely that a nascent field like ML can be subject to overwhelming hype. And, this makes a community vulnerable to snake oil merchants who exploit the enthusiasm of individuals with their pump and dump schemes. So it is crucial to moderate the community frequently and disconnect the malicious players who with their ignorance would weaken the links of budding an ML community.
