Cecilia (CC) Chiderski, a Solution Architect at CLD Partners, enjoys learning and creating innovative solutions for employees at professional service organizations to improve the user experience. We dive into CC’s journey learning Agentforce to then use those skills to build Proof of Concepts (POCs) and deployments for customers.
“By using Agentforce more, I was able to see how Salesforce is allowing me to create workflows between Salesforce, Slack, and Agentforce for users. To me, this is the Nirvana of any process!” CC shares their insights from building and deploying Agentforce solutions for professional service organizations and how to ensure successful client adoption.
What’s a typical day look like for you?
Well, I need my coffee! With my coffee in hand, I will catch up on any overnight emails that came through. I get a lot of the community emails—not just Agentforce—a lot of the Salesforce ones, like the Flow community or even like the Partner community. There are tons of emails that I receive and I scan through those. If there is anything that I’m like, “Oh, this is interesting,” I pin it and put it on the side for reading later.
After that, it really depends! It depends on the meetings of the day and on the projects that I’m working on. Every day is super different. Lately, I’ve been spending a lot of time wrapping up design with a pretty big implementation, which means I’m talking with the client, finalizing requirements and things like that. At another client, we are kicking off an Agentforce POC, so that’s also part of the routine. I’m getting pulled into more and more Agentforce POCs. I’m also getting ready for a go-live with another client. And finally, we are running an internal Agentforce hackathon, so that’s also part of the day lately. I’m supporting all of the teams, advising them, and providing them with whatever they need.
With over seven years working in the Salesforce ecosystem, how have you approached learning and staying up-to-date with new products like Agentforce?
It’s certainly a challenge. Trailhead, for sure. There is no question about that. That’s at the top of the list to start. Then, after that, I go to the amazing communities and use them! By reading, that’s really when I’m able to dig into new features, releases, events. The communities are a great resource to stay up-to-date on anything new, because community leaders and members share webinars, implementations, and more to help members.
The releases group is another big one. People put PowerPoints, and there are resources out there, it becomes a matter of constantly looking at them. In the Agentblazer Community on Slack, the broadcast channel for Agentforce release notes helps me a lot. I love that channel!
Ultimately, I’m more of a doer, so if you give me a sandbox and leave me alone for a couple of days, then I will likely learn what I’ve read about already.
What’s your biggest piece of advice for someone just starting off with Agentforce?
Trailhead is a great resource and it’s free.The Agentblazer Trails with the three statuses, allow you to go at your own pace. Join the community—it’s another great free resource. Then, think about something that sounds cool to you, and try to do it. You will get stuck. Go to the community to ask your questions and keep trying it. Eventually, you will get there and love it.
Why did you lean into learning Agentforce early on?
This past year is when I felt that Agentforce really was in front of our eyes. The number of available resources that Salesforce put out there, we didn’t have that before. With more resources and when I was able to get my hands on it, I began to start seeing the potential for our customers.
By using Agentforce more, I was able to see how Salesforce is allowing me to create workflows between Salesforce, Slack, and Agentforce for users. To me, this is the Nirvana of any process!
What’s it been like to take your Agentforce knowledge that you’ve built up to help help others internally and people in the community?
I love it! I love it! I love every single one of the Agentforce use cases that we did that are on the CLD Partners website to check out. My hope is that people watching the videos I share on LinkedIn get their wheels spinning to help them generate Agentforce use case ideas. It’s really a learning process. You have to be more creative sometimes with your solutions and that’s what I think our use cases are highlighting for people. That’s why we like to share them on social media and our website.
I hope people will react with “Oh, I can do that!” or “Is the agent really doing that?” to really get them thinking.
An example we shared was a status report agent. I wanted people to be saying, “Wait, what, the agent is reading all that and it’s being done in Flow?” This helps reduce the complexity and lower the entry barrier, since everybody knows Flow and is already using it. I love putting Agentforce use cases out there for whoever can benefit from it, go for it!
For your internal Agentforce hackathon, what’s it been like to be a mentor to your peers? What’s been the most exciting part of the hackathon for you so far?
The hackathon has been super exciting! It’s funny because one of my clients just finished their hackathon, and I helped them with some of the teams. When we decided to do one here at CLD Partners, I was super excited for it! It’s been really fascinating to listen to the teams coming up with their ideas. Some ideas stem from the things that they see at client organizations and are good problems to solve. We’ve had some great discussions.
Particularly, it’s really interesting the way they start by differentiating what’s really a good Agentforce use case versus using Apex or Flow. I think that it’s a super important part of the learning process.
Why did you lean into building employee-facing agent POCs to help people that you’ve shared in the Agentblazer Community on Slack and on LinkedIn?
Well, we do a lot of work with professional services organizations. Now, when you look at the cases that everybody’s working on or talking about, they’re typically Sales and Marketing ones. A lot of these are the ones that organizations, including Salesforce, can do well right out-of-the-box.
The reality that we see is a different set of use cases. We’re seeing services teams that have to deliver something for a client of a Professional Services organization. Here, you’ll have a project manager dealing with SOWs that they need to read and they need to digest all that information. They need to identify what’s in scope and out of scope, as well as the risks, issues, actions, project status reports, timecard reporting, resource management, and assign the right resource, happening on the delivery side of the house that nobody has been talking about or working on to solve. And, those organizations are our customers!
Our customers are the ones that are now saying things like, “Hey, how do I solve this? How do I make this better?” And, that’s why I’ve been focusing on those types of use cases that can help consulting companies and professional service departments run better. In many cases, we’re helping the organizations improve a process that is already in use and adding Agentforce on top of it for an extra assist.
A major challenge in many cases, I see, is the challenge to unlearn all that we know and reimagine it in the frame of a new tool, Agentforce. A lot of the use cases for a service organization are trying to solve their unique problems that are very different from the ones in Marketing and Sales.
Now, we can rethink all of their workflows with a different perspective, using Agentforce and Flow. I’m a Flow fanatic! The fact that I can use Agentforce and Flow together, to me, I was like, “Yes, that’s all I need!”
What’s your approach to building an impactful Agentforce POC?
Let’s start with aligning on a POC. A POC to me is an insurance policy. You are literally buying an insurance policy with a proof of concept. By starting small, it’s like you are applying the principle of, “fail fast, fail cheap.” If it’s not working, move to the next one–learn something from that use case and then move to the next one.
I think it’s even more important when we are talking about Agentforce, because it’s new. And, we as humans can struggle with new things. We don’t like the unexpected. We don’t like change. I believe a POC brings down a lot of that anxiety when leaning into new technology. A POC is going to help determine if the use case is a good one for Agentforce. By starting with a POC for an Agentforce use case, you can confirm whether the technology supports the outcomes you’re looking for from it for your use case. If yes, then you can start expanding by adding onto the initial use case for more value. By this point, you already took a lot of the anxiety out and a lot of the risk out. From my experience, I think that the emotional part in Agentforce implementations is a really big piece of it that people cannot ignore.
By acknowledging the emotional side of an Agentforce implementation and not starting with any use case in particular, but by being thoughtful with your approach you remove the ambiguity around how AI works for you and keeps the focus on measurable outcomes. If the POC helps us narrow down the initial use case that will be high-impact, low-cost. By layering Agentforce on top of that, we’re able to make it simpler without shaking up everything at once. Personally, I pay close attention to the emotional side of an implementation for clients, because I know that can make or break the value of an implementation, not always the technology itself.
How do you help organizations through the change management piece of an Agentforce implementation?
On the change management piece of an implementation there are tons of things you can do to help people adopt the new technology. The non-negotiable thing, for me, is to involve the client team from day one. These are the people that will be your champions–you must identify them early, bring them into the POC development process, because they will be the ones that are going to show its value to their peers internally. You cannot impose new AI solutions on people, since they need to see its value and build trust with it. The people that are going to be using it must be involved early on in its development.
What have your technical learnings from your Agentforce deployments, including any that are from deployments with Agentforce and Slack?
The fact that you can deploy AI agents in Slack with minimal additional work with Agentforce is great! That opens up an entire new world. That was definitely a learning that has supported deployments for our customers of Agentforce.
The other one is there is a lot of iteration. It’s a reality that you will try something and it may or may not work. You should test it with multiple use cases to ensure it results in the intended outcome each time. I find it helpful to have somebody else test it, so it’s not just me working and testing it. When you start asking other people to test it for their use case, that’s when you’ll learn something is missing from it or more guardrails need to be added to it. Putting a lot of time and effort on the testing side is super critical–it’s not a technical thing, but that you built a solution to help someone with their process.
If you haven’t identified a use case with a clear set of KPIs, then how will you know if your solution was valuable or not. It needs to have a KPI, at least one that you can clearly measure, prior to building the POC. By aligning on a use case with a clear KPI, you will know whether your Agentforce solution was successful and should proceed to implementation.
Examples of AI agent KPIs that I’ve seen from customers are reduction in number of tickets created or number of status reports that are sent on time. It’s super important, especially with the rapid innovation of Agentforce. If your AI agent requires knowledge—establish your governance early. You need to put your governance in place with clear processes to ensure you have the right information with regularly updated information, if needed, and continues to work with each new release. The work really starts after you deploy the agent.
What made you choose the Salesforce ecosystem as a career path?
I worked in business operations for 20+ years. I really started my career in business operations. I was on the other side of the table–the customer side. As a customer, we were implementing Certinia which works on Salesforce. And, CLD Partners, which is where I am now, was one of the integration partners we were evaluating to do the implementation. They were one of the two or three that we were evaluating to help with it. In the middle of that project, the company that I was working for moved the headquarters from Ohio to San Diego and I was impacted by a layoff when they closed the Ohio office.
CLD Partners reached out and offered me a job to join their team at the right time. It was a no-brainer for me, because I got to know them during the presale process. And, I already knew the business side, which is the hardest part to learn. It has been cool to continue to learn about the technology space. I like my job so much, because I know the pain that our customers go through from my time once being on the same side of the table as them. Now, I know technology can help them solve various pains that they experience.
The crazy part of the story is that when I was a customer and evaluating an implementation partner for Certinia, we didn’t go with CLD Partners at first! The partner we chose for the implementation was so bad that we ended up going with CLD Partners at the end of the day.
Why did you join the Agentblazer Community on Slack and become active there?
I’ve been a member for a while now! I joined when I was going through the Agentblazer Statuses on Trailhead. I really didn’t know what to expect when I joined at first. Then, I met Joel Primack and some of the other awesome people there. I’m glad I joined to access the resources shared there. I’ve become active through sharing what I have learned or things that I think can be useful for somebody else on their Agentblazer journey. I think the #agent-showcase channel really gets people to think. Maybe it’s not the use case that will resonate with them, but the fact that you see something being used in a different way may inspire someone to build a solution differently. By showing people that you can go from Slack to Salesforce and start a Flow—it gets people to really think. I like writing articles that add value for others building with Agentforce to share my knowledge and perspectives.
I especially like when I post something, because there might be a lot of people from minority groups reading it. I like making an impact on those people—maybe there are other women or Latin people– reading it. There is a lot of fear sometimes or a feeling that people in a minority group need to do more. I enjoy being able to show others that they can learn it and get hands-on with cutting edge technology solutions like Agentforce.
What do you enjoy doing outside of work?
A lot–anything that involves family and friends. That’s what’s important. It’s what comes first for me. I enjoy exercising, since I feel like it’s therapy for me. I love cooking and taking care of my plants. I do have a lot of plants and often get them on clearance to try to save them. The problem is that I do save them, so now if you come to my house there are all of these plants that I saved from the clearance section.
And, once in a while, I will enroll in a random class like cake decoration, dressmaking, or bartending. Seriously, I took them all! I’m interested in learning new things, so if I see the opportunity to try something new–I will. I’m trying to find a woodworking class but the classes are not at a good time. I’m curious by nature.
CC’s top 3 Agentforce learnings
- Implementing Agentforce as part of your organizational AI strategy requires more than just activating features. It requires a strategic framework and cross-functional alignment to deploy safely, compliantly, and in ways that generate measurable business value. Before starting:
- Ensure that you have defined the executive governance
- Build a cross-functional AI team with technical, business, and chance management expertise
- Assess current organization readiness & change management requirements
- Create a clear AI roadmap
- It is important to treat prompts as modular, reusable components instead of creating one-off prompt templates for each specific use case.
By applying the Separation of Concerns design principle from software engineering, teams can build general prompt templates that accept input parameters (like task and context) instead of hardcoding logic for every unique scenario, which helps avoid prompt proliferation and the technical debt that comes with it. This architectural approach not only makes prompt management easier but also ensures predictable outputs and simplifies governance and compliance as AI adoption grows across an organization.
- Going live with an agent is just the start… Continuous, active monitoring, evaluation, and adjustment are as important as the initial rollout to ensure agents are trusted, effective, and scalable. Keep in mind that new data, edge cases, and even changes to business processes WILL impact your agent over time. How to do this, though? Early on in the implementation, get familiar with Agentforce Observability tools and plan your monitoring strategy. These tools are designed to provide ongoing insight into agent behavior, decision paths, performance trends, and health signals. This enables active monitoring of outcomes, detecting inaccuracies, and making adjustments as needed.
Observability is not optional. Remember that the moment your users stop getting accurate answers, you lose momentum and credibility.
Want to meet more Agentblazers like CC?
Check back soon for more stories and join the community for the latest Agentforce content.


