AI for FinOps It’s the New Standard

In this episode of the Phenos Weekly podcast, Ido Codler, CPO of Belanor, and Damian Monafo discuss the evolving role of AI in Phops, emphasizing the importance of AI for Phops over Phops for AI. They highlight the potential of AI to automate routine tasks, enhance efficiency, and transform workflows, ultimately making Phops teams more central within organizations. Codler notes the need for AI to augment rather than replace human decision-making, stressing the importance of contextualizing data for actionable insights. The conversation also touches on the future of Phops, predicting a shift towards measuring cost efficiency in terms of business value rather than absolute spending.

• AI is anticipated to be a dominant theme for the decade, significantly influencing various industries, including Phops.
• There is a distinction between Phops for AI (managing AI-related operations) and AI for Phops (using AI to enhance Phops operations).
• AI for Phops is seen as a more immediate focus due to its potential to revolutionize workflows and enhance resource efficiency.

"AI is probably going to be like the word of the decade, not only of the year, like with the pace that things are going right now."

• AI's rapid development underscores its significant and long-term impact across sectors.

"AI for Phops is should like should be what we're tackling right now and not Phops for AI."

• Prioritizing AI for Phops is crucial as it offers immediate benefits in automating and improving Phops processes.

• Companies are in the early stages of integrating AI into Phops, with many still experimenting and understanding its potential.
• The pricing models for AI services are not yet standardized, complicating the management and forecasting of AI-related expenses.
• There is a need to wait for more stable models and practices before fully integrating Phops for AI.

"The way people will use AI in two years is probably going to be very different than how they use AI currently."

• The evolving nature of AI usage and pricing models presents challenges in establishing effective Phops strategies.

"Trying to create a process out of it too much is going to be incredibly difficult right now for many Phops teams."

• Establishing rigid Phops processes for AI is challenging due to the current volatility and lack of standardization.

• Current AI tools like Amazon Q and Copilot offer basic functionalities but lack advanced interactivity and integration with existing workflows.
• There is potential for AI to provide more contextualized and democratized data access, benefiting non-experts in Phops.

"What they're doing necessarily is super useful because it's super basic."

• Current AI tools are limited in their capabilities, offering only basic data representations.

"The ability to democratize, to contextualize, and do that even if those people are not really experts in what they're doing."

• AI's potential lies in making complex data accessible and understandable to non-experts, enhancing decision-making.

• The use of AI, particularly in automatically generated code and autonomous agents, poses risks such as errors and unintended actions.
• There is a need for careful management and oversight to prevent potential issues, such as data deletion or incorrect operations.

"The agents are super great but we have seen that they could make mistakes like they could be great but they could delete a database."

• Autonomous AI systems can be powerful but also pose significant risks if not properly managed.

"The trends that we're seeing with AI and let's say with the vibe of coding and people generally using very automatically generated code is extremely scary."

• The reliance on automatically generated AI outputs requires vigilant oversight to mitigate potential errors and risks.

• The potential risks of automation in code management include accidental over-provisioning of resources due to human error and the dangers of AI making autonomous decisions without human oversight.
• AI should serve as an augmentation tool for human tasks, not as an autonomous agent making decisions or executing actions without human validation.
• Effective AI integration involves gathering contextual data, engaging relevant stakeholders, and automating processes while maintaining human oversight.

"AI should augment what people currently do but not take the actions itself at the end of the day."

• The speaker emphasizes the role of AI as a supportive tool rather than an autonomous decision-maker, highlighting the importance of human oversight.

"The bottleneck is often that people first of all don't really trust the insight that they see."

• Trust in AI-generated insights is crucial; people need to verify data and understand the context before taking action.

"Automatically investigating integrating with all of those external systems getting this data and wherever it doesn't have the data sending a Slack message, requiring a response, opening a Jira ticket."

• AI can streamline processes by automating data collection and engaging stakeholders through communication tools, but human intervention remains necessary.

• AI-generated queries can lead to significant costs if not properly managed, as seen with costly SQL queries due to poor parameter settings.
• The prevalence of AI-generated content can result in increased instances of inefficiencies due to lack of expertise or oversight.
• The necessity for careful review and understanding of AI-generated outputs to prevent unnecessary expenses.

"We saw a huge spike in one of our customers for like a query that ran on some data lake within their infrastructure that basically costs like tens of thousands of dollars for one query."

• This example illustrates the financial impact of unoptimized AI-generated queries, underscoring the need for careful review and validation.

• Phenops (financial operations) shares similarities with cybersecurity, particularly in the integration and response to alerts and anomalies.
• Historically, Phenops lagged in cloud integration due to its origins in basic BI tools, but it is evolving to address complex cloud interactions.
• The shift towards integrating Phenops into engineering processes mirrors the evolution of cybersecurity practices.

"I think there are many parallels between PHOPS and security."

• The speaker draws a connection between Phenops and cybersecurity, suggesting that Phenops can learn from the evolution and integration seen in cybersecurity.

"Many of the vendors were created at a time that for example you couldn't even access the cost explorer in AWS but you had to download a bunch of like big Excel files."

• This reflects on the initial limitations of Phenops tools and the need for more integrated and advanced solutions.

• Alert fatigue, common in cybersecurity, is now a challenge in Phenops, with too many alerts leading to inefficiencies.
• AI can help by contextualizing alerts, reducing noise, and making recommendations more actionable and relevant.
• The importance of prioritizing alerts and using AI to identify the most critical issues amidst a flood of notifications.

"The noise of alerts and the fact that there are many things that are irrelevant. All of the tools throw you like a million tickets and you don't really know what to do."

• This highlights the challenge of alert fatigue, where excessive notifications can lead to important issues being overlooked.

"AI is incredibly good at like smelling what's not good enough like let's say you had an anomaly."

• AI's ability to filter and prioritize alerts can significantly improve efficiency and focus on critical issues.

• Budget alerts need to be set with context and correlation to time to avoid unnecessary noise and inefficiencies.
• Managing multiple accounts across a cloud infrastructure can lead to alert overload, necessitating smarter alert management strategies.
• The need for a strategic approach to setting and managing budget alerts to ensure they are meaningful and actionable.

"When you put budget alerts where do you put it like 10% 20% 50% you're getting all this all these nets you know that that doesn't give you anything first if you don't correlate it with time."

• This quote emphasizes the complexity of setting effective budget alerts and the need for context to make them meaningful.

• Monitoring every resource individually can lead to numerous false positives due to frequent changes like autoscaling, resulting in a lack of context and understanding of the broader impact.
• Focusing on higher-level alerts and understanding the root cause of changes can provide a more comprehensive view of incidents.
• Automating the analysis of alerts to identify interconnected mini-alerts can reduce noise and make alerts more actionable.

"If you're getting three different alerts, you're missing the full story."

• Highlighting the problem of fragmented alerts that fail to provide a complete picture of the situation.

"Doing alerts on the higher-level things makes more sense if you can once you get an alert, explain it really really well and find the root cause."

• Emphasizing the importance of high-level alerts that allow for a deeper understanding and investigation of root causes.

• AI will automate many menial tasks currently handled by FinOps teams, freeing up time for strategic activities.
• Automated reporting and trend analysis will streamline processes, saving time in data presentation and decision-making.
• AI will not reduce the need for FinOps professionals but will enhance their ability to focus on high-impact questions and strategic decisions.

"There are so many like uh menial tasks many FinOps teams do today that are going to be completely automated and that's going to be great."

• AI's potential to take over routine tasks, allowing teams to focus on more valuable work.

"Unlocking that possibilities for those people to ask much more higher-level questions."

• AI's role in enabling FinOps professionals to engage in strategic thinking and decision-making.

• AI is not poised to replace humans but to augment their capabilities, making them more efficient and effective.
• The trust in AI and robots will take time to develop, and human skills will remain valuable for the foreseeable future.

"The AI at this point is not to replace us not even close but it's there to make us better to help us be more efficient."

• Reassuring the role of AI as a tool for enhancement rather than replacement.

"I think we have many years there to go until you know AI and robots replace us."

• Acknowledging the long timeline before AI and robots can fully replace human roles.

• Effective communication and debating skills are crucial for gaining executive buy-in and influencing decision-making in FinOps.
• Listening and understanding the other party's perspective are key components of successful communication and debate.
• Engaging with the main points of contention and addressing them directly can lead to more productive discussions and outcomes.

"People usually when they think about debating they think about like people who like to argue. I think actually it's quite the opposite. It's more about people who like to listen."

• Debating is more about understanding and engaging with others than simply arguing.

"Finding out the main like points of contention and then talking specifically about them and finding out what did exactly the other person say about that and what would change their mind."

• The strategy of focusing on key issues and addressing them to facilitate effective communication and persuasion.

• Effective communication is crucial in PHOPS to avoid misunderstandings and disagreements between teams, such as engineering and finance.
• Establishing clear accountability across the organization helps streamline communication and decision-making.
• One-on-one agreements with stakeholders before larger meetings can prevent conflicts and facilitate consensus.

"The key thing in many of those things is finding out the incentive structures and finding out what drives people to do things."

• Understanding what motivates individuals within teams can improve collaboration and decision-making processes.

"Make sure you get to a one-on-one agreement because getting to one-on-one agreement is like a lot easier usually."

• Pre-meeting agreements with key stakeholders can simplify larger discussions and prevent prolonged disagreements.

• People skills are essential for PHOPS professionals due to the need to interact with diverse teams and personalities.
• Empathy and understanding others' motivations can facilitate smoother project execution and collaboration.

"First of all, you need to be a people person, a people guide, you know, you need to get along because eventually, you need to communicate with so many different teams."

• Interpersonal skills are prioritized over technical skills in PHOPS roles due to the collaborative nature of the work.

"Most people want to do good and most people want to save money... It's about what they're being measured on and what can they also use for their own goals."

• Recognizing that team members have positive intentions and personal goals can aid in building effective working relationships.

• The focus in PHOPS is shifting from absolute costs to cost per value, emphasizing efficiency and business value.
• This change is driven by evolving pricing models and the need to align costs with business outcomes.

"The unitless dollar is going to die... It's more going to be about the cost per value."

• The traditional focus on total expenditure is being replaced by evaluating the value derived per cost unit.

"The conversation only makes sense when it plays to unit. So I think people and PHOPS and leadership are less going to talk about how much we spend but more about are we doing that efficiently?"

• Efficiency and value per dollar are becoming central metrics for evaluating PHOPS effectiveness.

• AI and automation are expected to enhance PHOPS by simplifying the linkage between service consumption and business value.
• The integration of AI could streamline unit economics, making it easier to relate costs to business outcomes.

"I hope AI can help on that, and automation can help on that."

• AI and automation are anticipated to play a significant role in advancing PHOPS practices by improving cost-value analysis.

"We need to be better at automating unit economics and to, you know, being able to link it."

• Automating the process of relating costs to business value is seen as a crucial development for the PHOPS ecosystem.