← Home
Partner Basecamp · Cohort 7

San Francisco

May 19-20, 2026
+32
Net Promoter Score
65
Day 1 Responses
83%
Matched Pairs
92%
Day 2 Response Rate
Programme Satisfaction
Positive NPS of +32 — promoters outweigh detractors with moderate passive presence.
+32NPS · n=60
53% Promoters25% Passives22% Detractors
95% CI: +11 → +52  ·  True NPS lies within this range with 95% confidence (n=60 respondents)
Confidence Arc
Strong end-of-Day-1 build confidence — cohort leaves Day 1 ready to apply Claude in client work (mean 4.1/5).
4.1end-of-D1 build / 5
4.3D2 Design / 5
4.3D2 Commercial / 5
4.2D2 Build / 5
Audience
65 participants across 6 organisations — Transformation Lead majority with applying in practice the most common AI experience level.
Top Organisations
Accenture (17) Deloitte (15) PwC (15) Cognizant (9) Infosys (7)
Voice of Participant
What participants said — dominant theme from each NPS segment.
Promoters
Participants valued the course as an effective baseline introduction with practical, hands-on learning that exposed them to Claude's latest capabilities.
Passives
Attendees wanted deeper technical guidance on building solutions and more discussion time to digest concepts rather than rapid exercise cycles.
Detractors
Participants criticized the workshop for being introductory copy-paste coding exercises better suited for self-study, lacking real-world problem-solving and conceptual depth for technical audiences.
Facilitator Notes
Gaps Filled & Course-Corrections
Day 1: Access to the fac site: I had to attempt 10 times before it let me get through, same for day 2. We need to fix this We need to continue to be more prescriptive with instructions when sending them off to do workshops/ mini hacks. For example: When sending them off to activity and timer on screen have the name of the activity/ file also displayed When they’re in an activity for 45 min +, at least 1-2 check-ins from the front of the room. We’re doing great at grazing through the room but this will help the room understand how far along they should be, or also ask questions that others in the room might be wondering. At certain points it felt as if the pax were wondering “what are we supposed to be doing” Telling them they should close out of extra programs/ widows due to wifi bandwidth, and will help keep their attention so they aren’t on other tasks Day 2: OBSERVATION: Pax felt ready to dive right in, they were helping on another and really leaned into asking Claude first FIX NEEDED: images in the content engineering section are way too small and dont enlarge when you zoom in, can we replace these? After lunch folks didn’t seem to be paying attention as much, running different screens ect. we need to get the energy going before we dive into content
Room Dynamics
Day 1: Energy seemed lower towards end of day, we belive this is because they have been hands on with their key boards and the room overall is a little quiet. This is not negative, but maybe we can find a way to make this feel more interactive, have more table lvl discussions or larger room discussion throughout that last chunk Day 2: People were able to jump right in with minimal tech issues.
For Next Time
Day 1: while they’re coming in to the room have the prework up on screen, prompt them to get into it if they haven’t already. will help reduce the amount of time spent getting them into get later
Survey & Data Quality
Survey & Data Quality
Day 1 responses65
Day 2 responses60 (92% of Day 1)
Matched pairs50 (83% of Day 2)
Orgs resolved65 of 65 respondents matched to named org
Day 2-only respondents10 (16.7%) — suppresses NPS by 8 points; no Day 1 data, excluded from persona & confidence analysis
Organisations
All Day 1 respondents · n=65
Accenture17 (26%)
Deloitte15 (23%)
PwC15 (23%)
Cognizant9 (14%)
Infosys7 (11%)
Bounteous2 (3%)
Function × Seniority
All Day 1 respondents · n=65
FunctionPractitionerSr PractitionerMgr / Sr MgrDirector+Partner / MDTotal
Engineering6762–21
Architecture1245–12
Business Leadership––26210
Project / Engmt44103122
Experience Profile
AI experience level · n=65 · % of row
By Persona
SegmentExploringApplyingDeliveringFrontiern
Transformation Lead19%69%9%3%32
Developer10%67%19%5%21
Architect8%17%58%17%12
By Seniority
SegmentExploringApplyingDeliveringFrontiern
Manager or Senior Manager18%59%18%5%22
Director, Senior Director, or Principal6%44%31%19%16
Senior practitioner (5–9 years)8%77%15%—13
Practitioner (0-4 years in role)18%64%18%—11
Partner, Managing Director, or Executive33%33%33%—3
How would you describe your current experience with AI tools prior to Basecamp?
Learning and exploring
9 (14%)
Applying in practice
38 (58%)
Delivering independently
14 (22%)
Operating at the frontier
4 (6%)
Prior to today, how much had you worked with Claude or the Anthropic API?
Not at all
4 (6%)
A little
35 (54%)
Regularly
26 (40%)
Did the depth land for this audience?
Technical depth perception · n=65 respondents · % of row
By Persona
SegmentToo basicAbout rightToo advancedn
Transformation Lead6%78%16%32
Developer5%95%—21
Architect33%67%—12
By AI Experience Level
SegmentToo basicAbout rightToo advancedn
Applying in practice5%87%8%38
Delivering independently21%79%—14
Learning and exploring—78%22%9
Operating at the frontier50%50%—4
Overall depth distribution
Too basic
7 (11%)
About right
53 (82%)
Too advanced
5 (8%)
Did the pace work across the room?
Session pace perception · n=65 respondents · % of row
By Persona
SegmentToo slowWell pacedToo fastn
Transformation Lead3%84%12%32
Developer—100%—21
Architect8%92%—12
By AI Experience Level
SegmentToo slowWell pacedToo fastn
Applying in practice3%92%5%38
Delivering independently—100%—14
Learning and exploring—78%22%9
Operating at the frontier25%75%—4
Overall pace distribution
Too slow — could have covered more
2 (3%)
Well paced
59 (91%)
Too fast — not enough time to apply
4 (6%)
How confident were participants to build with Claude after Day 1?
End-of-Day-1 confidence · "How confident are you in your ability to build a client solution using Claude?" · n=65 · 1–5
Mean end-of-D1 build confidence by Persona
Architect4.33/5 · n=12
Developer4.19/5 · n=21
Transformation Lead3.91/5 · n=32
Programme mean: 3.9/5 · 15 cohorts
Mean end-of-D1 build confidence by AI Experience Level
Operating at the frontier5.00/5 · n=4
Delivering independently4.57/5 · n=14
Applying in practice3.92/5 · n=38
Learning and exploring3.56/5 · n=9
Overall end-of-D1 build confidence distribution
1
0 (0%)
2
2 (3%)
3
14 (22%)
4
26 (40%)
5
23 (35%)
How relevant was today's content to your current role?
Content relevance rating · n=65 respondents · 1 = not relevant, 5 = highly relevant
Mean relevance by Persona
Architect4.42/5 · n=12
Transformation Lead4.03/5 · n=32
Developer3.95/5 · n=21
Mean relevance by AI Experience Level
Operating at the frontier4.75/5 · n=4
Delivering independently4.21/5 · n=14
Applying in practice4.03/5 · n=38
Learning and exploring3.78/5 · n=9
Overall relevance distribution
1
1 (2%)
2
3 (5%)
3
10 (15%)
4
27 (42%)
5
24 (37%)
How likely are you to recommend attending this programme to a colleague?
n=60 Day 2 respondents · 92% of Day 1
Score distribution · 0 = not at all likely · 10 = extremely likely
Cohort012345678910
Cohort 72563121022
■ Detractors 0–6■ Passives 7–8■ Promoters 9–10
NPS Breakdown
53%
25%
22%
NPS +32 (n=60)
Promoters (9–10)Passives (7–8)Detractors (0–6)
Programme mean: +42 · 15 cohorts
Promoters mean: 51%
Passives mean: 40%
Detractors mean: 9%
NPS by Segment
Organisations, personas, and experience levels · ≥4 respondents shown
By Organisation
Promoters (9–10)Passives (7–8)Detractors (0–6)
Accenture
73%
27%
0%
NPS +73 (n=11)
Deloitte
69%
23%
8%
NPS +62 (n=13)
Cognizant
29%
29%
43%
NPS -14 (n=7)
PwC
18%
18%
64%
NPS -45 (n=11)
Infosys
33%
33%
33%
NPS 0 (n=6)
By Persona
Promoters (9–10)Passives (7–8)Detractors (0–6)
Architect
50%
38%
12%
NPS +38 (n=8)
Transformation Lead
54%
17%
29%
NPS +25 (n=24)
Developer
44%
28%
28%
NPS +17 (n=18)
Programme means · 15 cohorts: Architect +52 · Developer +49 · Transformation Lead +37
By AI Experience Level
Promoters (9–10)Passives (7–8)Detractors (0–6)
Operating at the frontier
75%
25%
0%
NPS +75 (n=4)
Delivering independently
70%
20%
10%
NPS +60 (n=10)
Applying in practice
43%
23%
33%
NPS +10 (n=30)
Learning and exploring
33%
33%
33%
NPS 0 (n=6)
Programme means · 15 cohorts: Learning and exploring +48 · Applying in practice +47 · Delivering independently +40 · Operating at the frontier +32
What is the main reason for your score?
n=52 responses · organised by NPS segment
Promoters (score 9–10)· 26 responses
Participants valued the course as an effective baseline introduction with practical, hands-on learning that exposed them to Claude's latest capabilities.
10“Content was great. Presenters were knowledgeable and very helpful. Exercises focused on the content they just covered with time to experiment . Course length a d breaks were well times Thanks”
Passives (score 7–8)· 13 responses
Attendees wanted deeper technical guidance on building solutions and more discussion time to digest concepts rather than rapid exercise cycles.
8“In comparison with Microsoft AI-300 , the content of the course being taught is level above. While Microsoft AI-300 teaches me evaluations, tracing at each pipelines and the differential factors , Claude touches on evaluations but it doesn’t touch on what causes latency and throughput and each pipeline tasks.”
Detractors (score 0–6)· 13 responses
Participants criticized the workshop for being introductory copy-paste coding exercises better suited for self-study, lacking real-world problem-solving and conceptual depth for technical audiences.
4“I found the training helpful for a few takeaways / tips but not much more broadly. I understand the challenge in crafting the exercises but I thought they were too advanced for someone who comes in knowing nothing but too basic for someone who comes in having foundational knowledge. I found the Jupyter notebooks confusing to use, and the instructors weren't clear until speaking to them 1:1 that the preferred method was to run it directly in Claude Code or VS Code. Additionally, none of the instructions were written for the desktop Claude Code. The first exercise that had instructions in a site directly (https://claude-code-workshop.netlify.app/) was by far the best, and I would have loved if the other exercises were the same. The 02 and 03 exercises on Day 2 in particular, were not very helpful. For 02 especially, there was nothing actually to DO because it ran on its own and I didn't understand what the takeaway was supposed to be. Our table finished in just a few minutes and chatted instead because there was nothing else to do for that exercise. In general across the exercises, I didn't understand the right balance of what to ask Claude to do directly, versus what we should do ourselves. I liked that in the 04 exercise on Day 1 they told us more explicitly when to digest ourselves vs. when to use Claude. Because Claude could do the exercises itself, I felt like I didn't necessarily grasp the takeaways. I think the Agent Build at the end of Day 2 was one of the better exercises, but for all of them, there was a lot already built / a lot of code already written, which personally made it more confusing. Since I have less coding experience, I didn't know what to read / how to understand a good amount of what was in the Jupyter notebooks. There were parts to fill in that seemed to require Python knowledge or understanding of the data structure. While I was able to move forward by talking to Claude, it felt like it defeated the purpose of the exercise. I also really liked the evals conversation and exercise, and it was where I felt like I learned the most. However, it wasn't made clear how/where to implement the eval in a real world use case. Do you build a site? When does the eval check get run? It would have been helpful to connect back to use cases and when to literally put an eval into practice and how. Finally, this is super small and I'm sure unintentional, but I was doing an initial group share with a group of other women and one of the facilitators came to chat with us. She spoke to us with the assumption that we didn't know anything about Github/development, despite us not having shared that. It felt like because we were a group of women, it was assumed that we were on the lower end of experience, when in fact that was not true. Again, she shared helpful tips (which may have been shared with other groups as well), it just struck me as odd. It was also surprising to me, and the other participants I spoke to about this, that none of the facilitators worked at Anthropic. It felt like it was billed as an Anthropic-run training and the facilitators were purposefully vague about it. I think it would have been better if there was more transparency there! The facilitators were still helpful and knowledgeable, and Claude directly was able to do almost all troubleshooting anyways.”
End-of-Programme Confidence
n=60 respondents · 1 = not confident, 5 = very confident
How confident are you in your ability to design an evaluation for an AI solution?
12345Mean
——18%35%47%4.28
Programme mean: 4.1/5 · 15 cohorts
Leaving today, how comfortable are you advising clients on AI when questions arise?
12345Mean
——15%42%43%4.28
Programme mean: 4.2/5 · 15 cohorts
How confident are you in your ability to lead a conversation about Anthropic and Claude with a client?
12345Mean
—3%13%47%37%4.17
Programme mean: 4.2/5 · 15 cohorts
Confidence Trajectory by Persona
D1→D2 delta · apply-AI confidence · matched pairs only · n=50
Positive delta = confidence grew · negative = dropped
Architect (n=8)
+0.12
Developer (n=18)
-0.17
Transformation Lead (n=24)
+0.21
What one takeaway will you share with a colleague or client?
n=39 responses
Participants recognize Claude's capability but prioritize learning surrounding implementations like cost management and context engineering.
“Models are highly capable by themselves what matters are the surrounding implementations- cost management , evaluations , context engineering , model selection”
Participants value practical evaluation frameworks as essential knowledge for assessing and judging model responses in production.
“Using evals to judge responses”
Participants recognize that prompt engineering and evals are critical but frequently overlooked in agent and ecosystem implementations.
“The focus on prompt engineering and evals are even more important and often overlooked”
Participants gain practical understanding of how AI agents and ecosystem tools function to unlock diverse use cases.
“How Ai agents and other tools within the ecosystem work”
Participants shift focus from model selection toward application-layer engineering as the actual source of differentiation.
“Model is just commodity and we should focus more on application layer and engineering”
What will you build for yourself or apply at work in the next 30 days?
n=48 responses
Participants want to design, build, and deploy complete agentic systems rather than learn isolated concepts.
“Full agentic applications”
Consultants need practical Claude knowledge tied to cost modeling and token economics for client conversations.
“Mainly build using claude and also understand the cost and tokens around the models to have discussion with clients”
Participants seek ready-to-use tools and materials that accelerate delivery and enable sales of AI solutions.
“Delivery tools to help project teams and GTM materials to help us sell more work”
Consultants want to define custom agents with governance controls tailored to specific workflow applications.
“Build and define custom agents with improvised controls for my application”
Hands-on evaluation practice and prompt engineering workshops enabled participants to scale skill improvements systematically.
“The prompt engineering and eval workshops were the best.”
Partner Basecamp Survey Report · Cohort 7 · Generated by analyze.py v2.3
NPS Reason
n=52 responses · grouped by NPS segment
Promotersscore 9–10 · 26 responses
1Very helpful for a beginner10
2Great practical session10
3Learnt new things and hearing about others’ experiences with Claude really helps think out of the box.10
4Very informative10
5Actually useful contents10
6Eye opening , clarity of thoughts in positioning Claude to customers10
7New knowldgr10
8Learned new ways to fully get full power of Claude10
9The contents, instructors and hands on exercises were greatly managed10
10Very engaging and practical experience10
11Seamless execution of realtime application construct and deploy process using Claude10
12Great session10
13It was well taught, attentive to details and folks were always present to support individuals. I think everyone could benefit from this program regardless of their AI literacy.10
14hands on is valuable10
15Content was great. Presenters were knowledgeable and very helpful. Exercises focused on the content they just covered with time to experiment . Course length a d breaks were well times Thanks10
16We could deeply learn about agent development, especially in context engineering and evaluations10
17Solid for baselining people across the skill.9
18Great learning on latest features and capabilities9
19Day 1 was introductory. Can be split into a different session.9
20Staff was very helpful, I learned a lot9
21Content I have not seen elsewhere by Anthropic in training9
22Great hands-on exercises9
23Enjoyed the class9
24Great content and learning opportunity9
25Engaging content and the hands on practicals9
26It gives a holistic view kf clause enterprises and also build along using it which helps in curating real use cases9
Passivesscore 7–8 · 13 responses
1Na8
2Good course8
3Not as advanced as I would have liked.8
4It will be useful to learn new things8
5Good8
6I work primarily with engineers and think this would be a good course for them to get deeper into Claude. I do think that there were a few too many exercises and would rather have digested a bit more in between. I think it would be helpful to discuss the solutions more in depth8
7Lots of hands on session which helped to understand and playground with Claude code.8
8Good experience and able interaction with more people and share the knowledge8
9In comparison with Microsoft AI-300 , the content of the course being taught is level above. While Microsoft AI-300 teaches me evaluations, tracing at each pipelines and the differential factors , Claude touches on evaluations but it doesn’t touch on what causes latency and throughput and each pipeline tasks.8
10Good hands on session - but didn't cover the aspects of using AI and claude in specific in solving real world problems8
11Would like more build guidance instead of working off pre built plumbing7
12I learned a lot of things via this training, but Japanese people may struggle to learn because of English skill.7
13Content is good; Anthropic staff helpful7
Detractorsscore 0–6 · 13 responses
1The online modules are nice. The working modules in the workshop are something one can do on their own and need not be done in person. There was no big insight6
2Its is good for entry level with no claude code experience6
3The workshop could have been more real world practice. Like actually solving by real world problems.6
4I feel that a lot of the topics were not new and well known topics. If someone was new this would be good to learn the concept but someone who knows the concepts it was very much repetitive. Also think there could be more talk and discussion then sitting and doing individual hands on labs.6
5I think this felt more of an introduction to how to use Claude and while hit on technical topics the way it was delivered felt more like copy and paste or execute this juniper notebook vs explaining the concepts and why we would do certain things.6
6While informative, I think it was somewhat narrowly focused on the engineering aspect. Many of us are not engineers, and might benefit from less technical building and more around what is possible. Clients are not asking us about how to troubleshoot their agents, but rather how can they use AI to help to begin with.6
7There is an opportunity to make the training facilitate more dialogue and discussion. This one focused a lot on heads down coding that can be pre-work/virtual work.5
8Understanding of python code is required to make this session useful. It's not really for non technical people5
9Training is not hands on or intuitive for non-tech folks. There have to be atleast one hands on to ensure the business and sales folks can harness the power of agentification in code form in Claude code or vscode5
10Session should have been focused more on particular role instead of focusing multiple roles, I feel not completely focus on presentation to client , not completely developer oriented5
11The early demos were very complex and hard to follow when jumping in. Additionally, they required additional set up steps. Several demos and workshops labs didn’t work immediately and required Claude to troubleshoot and fix. This took the focus away from the specific skills being taught in each lab5
12I found the training helpful for a few takeaways / tips but not much more broadly. I understand the challenge in crafting the exercises but I thought they were too advanced for someone who comes in knowing nothing but too basic for someone who comes in having foundational knowledge. I found the Jupyter notebooks confusing to use, and the instructors weren't clear until speaking to them 1:1 that the preferred method was to run it directly in Claude Code or VS Code. Additionally, none of the instructions were written for the desktop Claude Code. The first exercise that had instructions in a site directly (https://claude-code-workshop.netlify.app/) was by far the best, and I would have loved if the other exercises were the same. The 02 and 03 exercises on Day 2 in particular, were not very helpful. For 02 especially, there was nothing actually to DO because it ran on its own and I didn't understand what the takeaway was supposed to be. Our table finished in just a few minutes and chatted instead because there was nothing else to do for that exercise. In general across the exercises, I didn't understand the right balance of what to ask Claude to do directly, versus what we should do ourselves. I liked that in the 04 exercise on Day 1 they told us more explicitly when to digest ourselves vs. when to use Claude. Because Claude could do the exercises itself, I felt like I didn't necessarily grasp the takeaways. I think the Agent Build at the end of Day 2 was one of the better exercises, but for all of them, there was a lot already built / a lot of code already written, which personally made it more confusing. Since I have less coding experience, I didn't know what to read / how to understand a good amount of what was in the Jupyter notebooks. There were parts to fill in that seemed to require Python knowledge or understanding of the data structure. While I was able to move forward by talking to Claude, it felt like it defeated the purpose of the exercise. I also really liked the evals conversation and exercise, and it was where I felt like I learned the most. However, it wasn't made clear how/where to implement the eval in a real world use case. Do you build a site? When does the eval check get run? It would have been helpful to connect back to use cases and when to literally put an eval into practice and how. Finally, this is super small and I'm sure unintentional, but I was doing an initial group share with a group of other women and one of the facilitators came to chat with us. She spoke to us with the assumption that we didn't know anything about Github/development, despite us not having shared that. It felt like because we were a group of women, it was assumed that we were on the lower end of experience, when in fact that was not true. Again, she shared helpful tips (which may have been shared with other groups as well), it just struck me as odd. It was also surprising to me, and the other participants I spoke to about this, that none of the facilitators worked at Anthropic. It felt like it was billed as an Anthropic-run training and the facilitators were purposefully vague about it. I think it would have been better if there was more transparency there! The facilitators were still helpful and knowledgeable, and Claude directly was able to do almost all troubleshooting anyways.4
13There were a great number of flaws in the notebooks, little instruction, and we did not receive accurate setup instructions prior to being onsite. The main response from facilitators was “ask Claude”. I’m not certain in that case the purpose of being onsite. I would suggest more robust setup instructions including all the dependencies we need. I would also recommend reviewing all trainings and fixing the major issues. Some had issues in the very first cell because someone forgot to install the necessary dependencies before importing a module. There was also far too many variables involved. If we used Claude through VS Code or CodeSpaces as suggested, we’d run into rate limit issues from Git. Using the native Claude Code application didn’t have the same views of directories. We were supposed to set up Claude Cowork but never did anything with it. There is apparently a certification involved with this and we got no information about it at any point across the past two days.4
Most Valuable
What one takeaway will you share with a colleague or client? · n=39 responses
1Models are highly capable by themselves what matters are the surrounding implementations- cost management , evaluations , context engineering , model selection
2Na
3Evals learnings
4Keep learning and stay updated
5How Ai agents and other tools within the ecosystem work
6It’s not as scary as I thought
7The only limit is our imagination
8AI
9Claude can be used in multiple ways depending on who is needing it
10Start buying Claude personal account
11Using evals to judge responses
12The importance of Evals
13Learn how to plan before implementing No swag ;( Please provide channel of communication.
14The models are production ready and let’s go!
15Evals
16Latency and cost are a big talking point
17The focus on prompt engineering and evals are even more important and often overlooked
18Claude for all
19Great course for intro to Claude
20Model is just commodity and we should focus more on application layer and engineering
21Learn it by experiencing it
22Claude is amazing, just ask if you’re not sure. No question is too silly
23Good
24Agent building
25Better preparation is recommended and some technical know how
26Please get a discord chat so we can share messages, output, results, etc. Communication would be a great way of sharing
27Very interesting ways to utilize a tool like this.
28Learn more on AI
29Agent harness, inference troubleshooting. One of my favorite and new topics. Thank you.
30Good one, wanted more of detailed information
31You can make a flexible agent
32How it will benefit for the team.
33Everyone needs to now be able to build using AI and claude is where the buck stops. The capabilities are amazing and we just touched the tip of iceberg here
34How we can leverage Claude to produce faster with better cost optimization
35Methodology of eval
36・The importance and best practice of context engineering ・how to evaluate agents, like code, llm as a judge or human fb
37Na
38Using evals! Deterministic vs LLM evals
39Other feedback: Multiple people on my team have dietary restrictions. When we had our alliance reach out, we were told that you wouldn’t be taking dietary restrictions because there were a variety of food options in the Ferry Building. It sounded as if we’d be eating from places in the Ferry Building, not that you were catering and instead meant “we are not taking anyone’s dietary restrictions into account and you’ll have to buy your own food if you have them”. Myself and another team member were not able to eat breakfast or lunch the first day at all. It was highly unpleasant and we would have made other arrangements if it had been clear you do not make accommodations.
30-Day Intentions
What will you build for yourself or apply at work? · n=48 responses
11. Build - design and implement agentic system for solutions 2. Reimagine the BPO operating model
2Developer prod
3GRC tools
4Many agents
5Yes
6A local knowledge graph that can pull content together from various historical decks.
7Full agentic applications
8Custom apps
9poc for client.
10I will build a dashboard for a client to help decision making.
11Create RFP moderater
12Continue to iterate on current projects and client accelerators
13Multi-agent architecture
14Agent to help with daily task
15I’ll develop agents to help my tasks better and efficient
16Converting user story to agentic tasks
17Multiple Projects
18Personal website to show professional journey and interests
19Build a marketing platform run by a team of agents
20A dashboard for my work which will help in taking decisions on a day to day basis.
21Operations analytics dashboard
22Applying AI agents into workflows
23Evaluations
24Build and define custom agents with improvised controls for my application
25Looking to build a personal assistant agent and hoping to be able to leverage co-work when available on enterprise license.
26Internal utilities - couple of agents to do my day to day job
27Identify product transformation cross workstream Claude usage
28Delivery tools to help project teams and GTM materials to help us sell more work
29Agents
30Build agents for Pharma company
31Agents
32I plan to build agents for regular tasks , RFP questions, RFP solutions, that help me become more productive
33So many things. Building 5 different ideas now, why wait
34Migration tool for integration solutions
35Continue to use AI skills for POC and Proposals
36We’re building a firm-wide Skills marketplace that works downstream and upstream. What I’m excited to work on is implementing Evals against those skills and improve them at scale. One step at a time.
37Revisions to our current multi agent tool used internally to help us deliver to our clients.
38Presentation, building smart agents for detections.
39Will make the agent for myself to brainstorm
40Service now tickets automatic response and the resolution notes updation
41Mainly build using claude and also understand the cost and tokens around the models to have discussion with clients
42I’m working on an energy trading app. This course will help me finish faster so I can demo to our client faster and provide additional context on the approach using AI
43I will try to effort to make my company's enablement higher.
44Multi agent chat-based application for development and knowledge search with Strands SDK and Claude on bedrock
45Na
46Automate community building tasks
47I want to implement a routine (this was briefly shared in the "What's New" section and this is a feature I am super excited for) based on our Github tickets.
48I did appreciate that it got me working in code and a bit of time to play around. After the issues with setup, I did understand how to use it better. The prompt engineering and eval workshops were the best.