Guest post by Ganesh, Founder AiHello
1. Why are you building this startup? What does it do?
I am a long time online seller of physical goods primarily on Amazon and also on Ebay and Shopify. I noticed that most small medium sellers sell either by guessing or their gut feeling and mostly it’s a hit and miss game. Manual research of competition, pricing etc is time consuming and confusing. So I started building this predictive analysis tool which would use machine learning and help me make better decisions for my online ecommerce activities. Thus AiHello was born. AiHello analysis products, orders and competitors to suggest the best decisions for you.
2. What does a product marketing team need to understand from engineering about how to sell the product?
We have a cross-functional team where the marketing team provides a list of features which are derived from surveys and customer feedback. The list is arranged by priority and sent to the engineering/development team. The development team then discuss internally the efforts required for each feature and then finally have a discussion with the marketing team about the efforts. The priorities are then rearranged based on effort required. The product marketing team need to understand and trust that the engineering team have a good idea on the approximate effort required to develop a new feature. They will have to re-prioritize all the features based on their interactions with the engineering team.
3. What impact or legacy do you hope to make in the market and in the business world?
We hope to be the first and largest to use Machine Learning for small and medium companies which will help them make better decisions for selling their goods.
4. What advice do you have for anyone wanting to start a business in your country?
Australia and in particular Melbourne is just warming up to startups. You need to have patience and spend more time online trying to network with other startup founders & investors in order to gain exposure. There are research grants by the government if you are working on research project (like AI) that offers grants and tax rebates. Also, I would avoid copying startups from the rest of the world that has not launched in Australia with the intention of launching in Australia. Instead work towards starting unique in Australia targeting uniquely Australian consumers or launch globally with unique products.
5. How do you work with investors, or do you currently use any outside money at all?
Currently, we do not use any outside money as we are earning revenue from our users. We reinvest the revenue to help us grow and scale up.
6. Why would an entrepreneur turn to Microsoft for help in building scale, a team, or using software?
We have been an Azure user for a long time and Azure has come a long way in being a stable, enterprise ready platform. It’s very easy to build virtual machines on Azure and quickly scale them up as the situation demands.
Challenges - Working on Machine Learning and Deep Learning challenged our skills to the limits. The primary factor was that scant information was available online and whatever information we could find was scattered across for different programming languages and different libraries
Technology - We use Microsoft Azure Linux VM as our primary OS. Setting up a clustered load balancing network was very easy. We opted for Apache Spark and Spark ML for our Big Data analysis and DeepLearning4j for our Deep Learning functionalities.