Insights and tips about the right way to prepare for a successful transition
As Artificial Intelligence is becoming an increasing number of popular, more firms and teams want to begin or increase leveraging it. Due to that, many job positions are appearing or gaining importance available in the market. A superb example is the figure of Machine Learning / Artificial Intelligence Product Manager.
In my case, I transitioned from a Data Scientist role right into a Machine Learning Product Manager role over two years ago. During this time, I even have been capable of see a relentless increase in job offers related to this position, blog posts and talks discussing it, and lots of people considering a transition or gaining interest in it. I even have also been able to substantiate my passion for this role and the way much I enjoy my day-to-day work, responsibilities, and value I can bring to the team and company.
The role of AI / ML PM continues to be quite vague and evolves almost as fast as state-of-the-art AI. Although many product teams have gotten relatively autonomous using AI due to plug-in solutions and GenAI APIs, I’ll give attention to the role of AI / ML PMs working in core ML teams. These teams are often formed by Data Scientists, Machine Learning Engineers, and Research Scientists, and along with other roles are involved in solutions where GenAI through an API won’t be enough (traditional ML use cases, need of LLMs tremendous tuning, specific in-house use cases, ML as a service products…). For an illustrative example of such a team, you may check one among my previous posts “Working in a multidisciplinary Machine Learning team to bring value to our users”.
On this blog post, we are going to cover the fundamental skills and knowledge which can be needed for this position, the right way to get there, and learnings and suggestions based on what worked for me on this transition.
There are a lot of mandatory skills and knowledge needed to succeed as an ML / AI PM, but a very powerful ones will be divided into 4 groups: product strategy, product delivery, influencing, and tech fluency. Let’s deep dive into each group to further understand what each skill set means and the right way to get them.
Product Strategy
Product strategy is about understanding users and their pains, identifying the appropriate problems and opportunities, and prioritizing them based on quantitative and qualitative evidence.
As a former Data Scientist, for me this meant falling in love with the issue and user pain to resolve and never a lot with the precise solution, and occupied with where we will bring more value to our users as an alternative of where to use this cool latest AI model. I even have found it key to have a transparent understanding of OKRs (Objective Key Results) and to care in regards to the final impact of the initiatives (delivering outcomes as an alternative of outputs).
Product Managers must prioritize tasks and initiatives, so I’ve learned the importance of balancing effort vs. reward for every initiative and ensuring this influences decisions on what and the right way to construct solutions (e.g. considering the project management triangle – scope, quality, time). Initiatives succeed if they’re able to tackle the 4 big product risks: value, usability, feasibility, and business viability.
An important resources I used to study Product Strategy are:
- Good vs bad product manager, by Ben Horowitz.
- The reference book that everybody really useful to me and that I now recommend to any aspiring PM is “Inspired: The best way to create tech products customers love”, by Marty Cagan.
- One other book and creator that helped me catch up with to user space and user problems is “Continuous Discovery Habits: Discover Products that Create Customer Value and Business Value”, by Teresa Torres.
Product Delivery
Product Delivery is about with the ability to manage a team’s initiative to deliver value to the users efficiently.
I began by understanding the product feature phases (discovery, plan, design, implementation, test, launch, and iterations) and what each of them meant for me as a Data Scientist. Then followed with how value will be brought “efficiently”: starting small (through Minimum Viable Products and prototypes), delivering value fast by small steps, and iterations. To make sure initiatives move in the appropriate direction, I even have found it also key to repeatedly measure impact (e.g. through dashboards) and learn from quantitative and qualitative data, adapting next steps with insights and latest learnings.
To study Product Delivery, I’d recommend:
- A number of the previously shared resources (e.g. Inspired book) also cover the importance of MVP, prototyping and agile applied to Product Management. I also wrote a blog post on the right way to take into consideration MVPs and prototypes within the context of ML initiatives: When ML meets Product — Less is commonly more.
- Learning about agile and project management (for instance through this crash course), and about Jira or the project management tool utilized by your current company (with videos comparable to this crash course).
Influencing
Influencing is the flexibility to achieve trust, align with stakeholders and guide the team.
In comparison with the Data Scientist’s role, the day-to-day work as a PM changes completely: it isn’t any longer about coding, but about communicating, aligning, and (so much!) of meetings. Great communication and storytelling develop into key for this role, especially the flexibility to clarify complex ML topics to non technical people. It becomes also essential to maintain stakeholders informed, give visibility to the team’s exertions, and ensure alignment and buying on the long run direction of the team (proving how it can help tackle the largest challenges and opportunities, gaining trust). Finally, additionally it is essential to learn the right way to challenge, say no, act as an umbrella for the team, and sometimes deliver bad results or bad news.
The resources I’d recommend for this topic:
- The whole stakeholder mapping guide, Miro
- A must read book for any Data Scientist and likewise for any ML Product Manager is “Storytelling with data — A Data Visualization Guide for Business Professionals”, by Cole Nussbaumer Knaflic.
- To learn further about how as a Product Manager you may influence and empower the team, “EMPOWERED: Bizarre People, Extraordinary Products”, by Marty Cagan and Chris Jones.
Tech fluency
Tech fluency for an ML / AI PM, means knowledge and sensibility in Machine Learning, Responsible AI, Data usually, MLOPs, and Back End Engineering.
Your Data Science / Machine Learning / Artificial Intelligence background might be your strongest asset, be certain you leverage it! This information will mean you can talk in the identical language as Data Scientists, understand deeply and challenge the projects, have sensibility on what is feasible or easy and what isn’t, potential risks, dependencies, edge cases, and limitations.
As you’re going to lead products with an impact on users, including responsible AI awareness becomes paramount. Risks related to not taking this into consideration include ethical dilemmas, company fame, and legal issues (e.g. specific EU laws like GDPR or AI Act). In my case, I began with the course Practical Data Ethics, from Fast.ai.
General data fluency can be mandatory (probably you’ve it covered too): analytical pondering, being interested in data, understanding where data is stored, the right way to access it, importance of historical data… On top of that additionally it is essential to kow the right way to measure impact, the connection with business metrics and OKRs, and experimentation (a/b testing).
As your ML models will probably should be deployed as a way to reach a final impact on users, you may work with Machine Learning Engineers inside the team (or expert DS with model deployment knowledge). You’ll need to achieve sensibility about MLOPs: what it means to place a model in production, monitor it, and maintain it. In deeplearning.ai, you will discover a fantastic course on MLOPs (Machine Learning Engineering for Production Specialization).
Finally, it could possibly occur that your team also has Back End Engineers (normally coping with the mixing of the deployed model with the remainder of the platform). In my case, this was the technical field that was further away from my expertise, so I had to take a position a while learning and gaining sensibility about BE. In lots of firms, the technical interview for PM includes some BE related questions. Be sure that to get an summary of several engineering topics comparable to: CICD, staging vs production environments, Monolith vs MicroServices architectures (and PROs and CONTs of every setup), Pull Requests, APIs, event driven architectures….
We’ve covered the 4 most significant knowledge areas for an ML / AI PM (product strategy, product delivery, influencing and tech fluency), why they’re essential, and a few ideas on resources that may make it easier to achieve them.
Identical to in any profession progress, I discovered it key to define a plan, and share my short and mid term desires and expectations with managers and colleagues. Through this, I used to be capable of transition right into a PM role in the identical company where I used to be working as a Data Scientist. This made the transition much easier: I already knew the business, product, tech, ways of working, colleagues… I also searched for mentors and colleagues inside the company to whom I could ask questions, learn specific topics from and even practice for the PM interviews.
To organize for the interviews, I focused on changing my mindset: developing vs pondering whether to construct something or not, whether to launch something or not. I discovered BUS (Business, User, Solution) is a fantastic option to structure responses during interviews and implement this latest mindset there.
What I shared on this blog post can appear to be so much, but it surely really is way easier than learning python or understanding how back-propagation works. If you happen to are still unsure whether this role is for you or not, know that you may all the time give it a try, experiment, and judge to return to your previous role. Or perhaps, who knows, you find yourself loving being an ML / AI PM identical to I do!