Pros: The culture is full of insightful engineers who share knowledge freely and transparently.

Cons: Most of the company is based out of America, so you may require remote engagement with team members. If you prefer strictly in-person interactions, this may not be for you.

Pros: Everyone is very helpful. Great visibility from executives. A good place to develop and learn new skills. The engineering teams are very competent.

Cons: Culture can be improved. Not many people in the office.

Pros: Compensation and benefits (including lunch allowance) are above average by Toronto standards. The office here is a 20+ people outpost of the Santa Clara headquarters, focused on engineering hires and growing. Team culture is friendly, professional in conduct but casual in tone, and diverse, both here and throughout the entire company. The bar for new hires is medium-high, leading to a consistently competent team and respect among coworkers because I can assume that everyone in the company knows their stuff. Work hours are genuinely flexible. Leads are awesome, intelligent, listening to and supporting their teams, and make rational decisions. There are lots of interesting challenges to work on; people can move around tasks and teams according to their interests and strengths. Learning opportunities abound. For Toronto, we have bi-weekly Friday Happy Hours (in-house or at a pub) as well as a soccer team. Company finances are communicated transparently in quarterly meetings; the company is well-funded and on a solid growth trajectory. Engineers often have direct contact with clients/partners. Travel sometimes takes you to places such as Tokyo, Paris, or the Bay Area. The resulting product is useful and without ethical concerns or conflicts of interest for either party. Recent initiatives for internationalization and empowering external developers are exciting and a push in the right direction.

Cons: More work needs to be done in order to scale the company. Server builds are slow. CI infrastructure doesn't prevent breakage (although testing prevents breakage from getting pushed to production). Too much developer time is wasted. Code infrastructure for NLU is owned by or depends on a single person who isn't really looking for suggestions or contributions and is already overbooked. Machine Learning tech and capacity doesn't currently approach what Google can do. No company-sponsored open source contributions or conference visits.