← Home
Partner Basecamp · Cohort 11

San Francisco

May 26-27, 2026
+73
Net Promoter Score
49
Day 1 Responses
93%
Matched Pairs
84%
Day 2 Response Rate
Programme Satisfaction
Strong promoter majority — NPS +73 reflects broad programme endorsement.
+73NPS · n=41
73% Promoters27% Passives0% Detractors
95% CI: +60 → +87  ·  True NPS lies within this range with 95% confidence (n=41 respondents)
Confidence Arc
Moderate end-of-Day-1 build confidence — cohort is progressing; Day 2 has room to close remaining gaps (mean 3.7/5).
3.7end-of-D1 build / 5
4.1D2 Design / 5
4.2D2 Commercial / 5
4.2D2 Build / 5
Audience
49 participants across 7 organisations — Transformation Lead majority with applying in practice the most common AI experience level.
Top Organisations
Accenture (15) Deloitte (15) PwC (10) AlixPartners (5) Forgd.AI (2)
Voice of Participant
What participants said — dominant theme from each NPS segment.
Promoters
Participants valued hands-on exercises and code examples that made technical AI concepts accessible and reinforced learning through practical application.
Passives
Pacing and depth created tension: content-rich but time-constrained, with advanced technical material potentially overwhelming less experienced attendees.
Survey & Data Quality
Survey & Data Quality
Day 1 responses49
Day 2 responses41 (84% of Day 1)
Matched pairs38 (93% of Day 2)
Orgs resolved49 of 49 respondents matched to named org
Day 2-only respondents3 (7.3%) — suppresses NPS by 1 points; no Day 1 data, excluded from persona & confidence analysis
Organisations
All Day 1 respondents · n=49
Accenture15 (31%)
Deloitte15 (31%)
PwC10 (20%)
AlixPartners5 (10%)
Forgd.AI2 (4%)
Bounteous1 (2%)
Globant1 (2%)
Function × Seniority
All Day 1 respondents · n=49
FunctionPractitionerSr PractitionerMgr / Sr MgrDirector+Partner / MDTotal
Engineering2152–10
Architecture–351–9
Business Leadership––43310
Project / Engmt44101120
Experience Profile
AI experience level · n=49 · % of row
By Persona
SegmentExploringApplyingDeliveringFrontiern
Transformation Lead33%50%10%7%30
Developer20%50%30%—10
Architect11%56%11%22%9
By Seniority
SegmentExploringApplyingDeliveringFrontiern
Manager or Senior Manager33%42%17%8%24
Senior practitioner (5–9 years)25%62%12%—8
Director, Senior Director, or Principal14%29%29%29%7
Practitioner (0-4 years in role)17%83%——6
Partner, Managing Director, or Executive25%75%——4
How would you describe your current experience with AI tools prior to Basecamp?
Learning and exploring
13 (27%)
Applying in practice
25 (51%)
Delivering independently
7 (14%)
Operating at the frontier
4 (8%)
Prior to today, how much had you worked with Claude or the Anthropic API?
Not at all
2 (4%)
A little
29 (59%)
Regularly
18 (37%)
Did the depth land for this audience?
Technical depth perception · n=49 respondents · % of row
By Persona
SegmentToo basicAbout rightToo advancedn
Transformation Lead7%80%13%30
Developer10%90%—10
Architect33%67%—9
By AI Experience Level
SegmentToo basicAbout rightToo advancedn
Applying in practice8%80%12%25
Learning and exploring—92%8%13
Delivering independently14%86%—7
Operating at the frontier75%25%—4
Overall depth distribution
Too basic
6 (12%)
About right
39 (80%)
Too advanced
4 (8%)
Did the pace work across the room?
Session pace perception · n=49 respondents · % of row
By Persona
SegmentToo slowWell pacedToo fastn
Transformation Lead—80%20%30
Developer20%50%30%10
Architect—89%11%9
By AI Experience Level
SegmentToo slowWell pacedToo fastn
Applying in practice4%80%16%25
Learning and exploring—77%23%13
Delivering independently14%57%29%7
Operating at the frontier—75%25%4
Overall pace distribution
Too slow — could have covered more
2 (4%)
Well paced
37 (76%)
Too fast — not enough time to apply
10 (20%)
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=49 · 1–5
Mean end-of-D1 build confidence by Persona
Architect4.44/5 · n=9
Developer4.10/5 · n=10
Transformation Lead3.53/5 · n=30
Programme mean: 3.9/5 · 15 cohorts
Mean end-of-D1 build confidence by AI Experience Level
Applying in practice4.16/5 · n=25
Operating at the frontier4.00/5 · n=4
Delivering independently3.71/5 · n=7
Learning and exploring3.15/5 · n=13
Overall end-of-D1 build confidence distribution
1
1 (2%)
2
6 (12%)
3
9 (18%)
4
18 (37%)
5
15 (31%)
How relevant was today's content to your current role?
Content relevance rating · n=49 respondents · 1 = not relevant, 5 = highly relevant
Mean relevance by Persona
Developer4.30/5 · n=10
Transformation Lead4.27/5 · n=30
Architect4.22/5 · n=9
Mean relevance by AI Experience Level
Applying in practice4.44/5 · n=25
Learning and exploring4.31/5 · n=13
Delivering independently4.29/5 · n=7
Operating at the frontier3.00/5 · n=4
Overall relevance distribution
1
1 (2%)
2
1 (2%)
3
6 (12%)
4
17 (35%)
5
24 (49%)
How likely are you to recommend attending this programme to a colleague?
n=41 Day 2 respondents · 84% of Day 1
Score distribution · 0 = not at all likely · 10 = extremely likely
Cohort012345678910
Cohort 1156921
■ Detractors 0–6■ Passives 7–8■ Promoters 9–10
NPS Breakdown
73%
27%
0%
NPS +73 (n=41)
Promoters (9–10)Passives (7–8)Detractors (0–6)
Programme mean: +40 · 15 cohorts
Promoters mean: 50%
Passives mean: 40%
Detractors mean: 10%
NPS by Segment
Organisations, personas, and experience levels · ≥4 respondents shown
By Organisation
Promoters (9–10)Passives (7–8)Detractors (0–6)
Deloitte
82%
18%
0%
NPS +82 (n=11)
Accenture
75%
25%
0%
NPS +75 (n=12)
PwC
62%
38%
0%
NPS +62 (n=8)
AlixPartners
50%
50%
0%
NPS +50 (n=4)
By Persona
Promoters (9–10)Passives (7–8)Detractors (0–6)
Developer
78%
22%
0%
NPS +78 (n=9)
Transformation Lead
73%
27%
0%
NPS +73 (n=22)
Architect
71%
29%
0%
NPS +71 (n=7)
Programme means · 15 cohorts: Architect +50 · Developer +44 · Transformation Lead +34
By AI Experience Level
Promoters (9–10)Passives (7–8)Detractors (0–6)
Applying in practice
95%
5%
0%
NPS +95 (n=19)
Delivering independently
67%
33%
0%
NPS +67 (n=6)
Learning and exploring
60%
40%
0%
NPS +60 (n=10)
Programme means · 15 cohorts: Learning and exploring +43 · Applying in practice +41 · Delivering independently +40 · Operating at the frontier +37
What is the main reason for your score?
n=36 responses · organised by NPS segment
Promoters (score 9–10)· 26 responses
Participants valued hands-on exercises and code examples that made technical AI concepts accessible and reinforced learning through practical application.
10“I think this course is trying to tackle too much, but the course instructors weee able to weave it to provide value for non technical and technical. I wish we had gotten into some more content for day 2 on context engineering and other ideas, but most of the audience hasn’t used any Claude code so I understand why this was shorter”
Passives (score 7–8)· 10 responses
Pacing and depth created tension: content-rich but time-constrained, with advanced technical material potentially overwhelming less experienced attendees.
7“Great venue, plenty breaks, great content. But there's no way to really "get" the material without grappling with the content and in that regard it may have been too much content, or not enough time. Lots of the meat was at the end of the .ipynb which we often don't get enough time on”
End-of-Programme Confidence
n=41 respondents · 1 = not confident, 5 = very confident
How confident are you in your ability to design an evaluation for an AI solution?
12345Mean
—2%17%46%34%4.12
Programme mean: 4.1/5 · 15 cohorts
Leaving today, how comfortable are you advising clients on AI when questions arise?
12345Mean
—2%15%44%39%4.20
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
——15%51%34%4.20
Programme mean: 4.2/5 · 15 cohorts
Confidence Trajectory by Persona
D1→D2 delta · apply-AI confidence · matched pairs only · n=38
Positive delta = confidence grew · negative = dropped
Architect (n=7)
+0.14
Developer (n=9)
+0.00
Transformation Lead (n=22)
+0.77
What one takeaway will you share with a colleague or client?
n=23 responses
Participants recognized that context and harnesses matter as much as model selection when building effective AI solutions.
“Context is just as if not more important than the prompt itself”
Evaluation frameworks emerged as the single most valuable takeaway participants planned to implement with their teams.
“Evals evals evals”
Context engineering became the connective skill participants saw as essential for designing reliable AI agent systems.
“Context engineering”
Participants immediately shifted their approach to validate tool calls programmatically rather than relying solely on LLM judgment.
“I already shared with my team (during the evals session) how we need to adjust our approach -- utilizing programmatic checks on tool calls rather than relying on LLM as judge for the everything in an end-to-end fashion.”
Hands-on practice sessions proved essential for participants to translate conceptual learning into actionable team guidance.
“Practice”
What will you build for yourself or apply at work in the next 30 days?
n=28 responses
Participants want deeper technical exploration of agent architectures and frameworks beyond introductory coverage.
“Evals, but also exploring more of the agent harness”
Participants seek structured evaluation methodologies and additional testing routines for production-ready agent validation.
“Evals - more routines”
Participants value hands-on Claude exploration to translate capability demos into concrete applications for their own work.
“further playing with claude to identify use cases for my own day to day productivity and deliverable creation. passion projects to develop skills”
Participants connect technical learning to practical architectural decisions, from swarm applications to modernizing existing systems.
“Swarm agents web application”
Participants found concrete implementation examples, particularly managed agents, most grounded and actionable.
“Managed agents example”
Partner Basecamp Survey Report · Cohort 11 · Generated by analyze.py v2.3
NPS Reason
n=36 responses · grouped by NPS segment
Promotersscore 9–10 · 26 responses
1Interactive and fun10
2It was great content and challenges10
3Learned plenty of new things10
4Expanded insights and SOP for using. Good insights from others too10
5Hands on exercise10
6New learnings in an approachable way10
7Technical hands on experiences that reinforces the concepts10
8Need of the hour10
9Fundamental shift in how we approach or use the tool10
10So much learned! Got a lot I can bring to my clients10
11Cutting edge technology that is super important for our jobs. Hands on learning10
12Redefined how I work with Claude and reinforced using Claude for new work, which would have previously been out of reach.10
13Both learning new concepts and hands on building are useful10
14Everyone who wants to understand this brand new world has to attend to this kind of training to put hands on10
15I am very new to Claude code so this was a big learning experience for me and I feel more confident building agents10
16Incredible instructors and great agenda with lots of hands on learning, which is how I personally work best in with new content.10
17I think this course is trying to tackle too much, but the course instructors weee able to weave it to provide value for non technical and technical. I wish we had gotten into some more content for day 2 on context engineering and other ideas, but most of the audience hasn’t used any Claude code so I understand why this was shorter10
18I learned an intense amount in this brief time in a way that I am retaining & eager to apply as soon as I’m back home again. I’m already looking for aspects to upgrade with what I’ve learned — professionally & personally.10
19I learned a lot of valuable ways of thinking about using these tools, which is more important than just the techniques alone.10
20Hands on practice that could be applied to real life. Great networking opportunity.10
21Interactive and knowledgeable sessions.9
22Well thought exercises, code examples9
23I think it was very well run with a lot of helpful information.9
24It is extremely informative and offers a great space to dive deeper into these tools9
25Lot of new learning for me especially on the evals and the model selection criteria9
26Mixture of first hand work and slides9
Passivesscore 7–8 · 10 responses
1Program covered a lot of ground in a span of 2 days.8
2Very good way to get started on agentic ai8
3Can be too technical for others who aren’t tinkering8
4Good labs.8
5Very good general approach to learning products offered by Claude could use a bit more generalist knowledge for some colleagues8
6Varies in colleague experience and focus7
7Beautiful facility, great facilitators7
8Good info, good practice. Some messaging felt token heavy (just spend more :) )7
9Great venue, plenty breaks, great content. But there's no way to really "get" the material without grappling with the content and in that regard it may have been too much content, or not enough time. Lots of the meat was at the end of the .ipynb which we often don't get enough time on7
10I’d like to see more about the last mile of production ready deployment7
Most Valuable
What one takeaway will you share with a colleague or client? · n=23 responses
1Evals
2Context is just as if not more important than the prompt itself
3Harnesses are just as if not more important than the model
4Practice
5Context engineering
6Use Claude to solution itself when you get stuck!
7Eval framework, context engineering
8Evals evals evals
9Evals & planning mode
10Take the training
11It's not the model fault !
12Different way of thinking - getting out of search engine mode
13The reasoning power of LLMs and how to manage context
14Claude Code for debugging
15Al the material, could be create to access to the material you share on screens
16Use Claude for troubleshooting as much as possible
17We all feel behind. The goal is to just start
18Flexibility of Claude and different ways it can be use cased for faster processes
19The importance of evals!
20I already shared with my team (during the evals session) how we need to adjust our approach -- utilizing programmatic checks on tool calls rather than relying on LLM as judge for the everything in an end-to-end fashion.
21Don't shy away from Claude Code. Learn from it.
22Claude Desktop is not a production delivery mechanism
23How critical evals are
30-Day Intentions
What will you build for yourself or apply at work? · n=28 responses
1Claude demos
2More agents!
3Evals, but also exploring more of the agent harness
4CSM agent
5Managed agents example
6New skills to transform the way I work
7Multi agent systems to solve one of our day to day problems
8further playing with claude to identify use cases for my own day to day productivity and deliverable creation. passion projects to develop skills
9Swarm agents web application
10I will be setting up a few agents to accelerate my work
11The RFP agent and Onboarding agent
12Agentic AI solutions for modernization
13Not sure yet. Some ATS integration smarts
14Evals
15Evals - more routines
16LinkedIn posting agent!
17Agent to help team members with db queries
18More dashboard views
19Evals, memory, and context window
20Eval!
21For myself, I hope to build a product evaluator for baby essentials as I’m expecting this fall.
22Lots of fun things
23Will try to build skill files for agents to do audit testing
24More agents! For onboarding & training ongoing.
25Better evals, better team usage of Claude
26Re-design Java Spring web application to add container deployment
27Analytics agent
28I want to use cowork