← Home
Partner Basecamp · Cohort 35

San Francisco

October 22–23, 2026
+69
Net Promoter Score
43
Day 1 Responses
100%
Matched Pairs
91%
Day 2 Response Rate
Programme Satisfaction
Strong promoter majority — NPS +69 reflects broad programme endorsement.
+69NPS · n=39
69% Promoters31% Passives0% Detractors
95% CI: +55 → +84  ·  True NPS lies within this range with 95% confidence (n=39 respondents)
Confidence Arc
Strong end-of-Day-1 build confidence — cohort leaves Day 1 ready to apply Claude in client work (mean 4.4/5).
4.4end-of-D1 build / 5
4.4D2 Design / 5
4.4D2 Commercial / 5
4.4D2 Build / 5
Audience
43 participants across 17 organisations — Architect majority with applying in practice the most common AI experience level.
Top Organisations
IBM (8) Cognizant (7) Deloitte (6) PwC (3) Salesforce (3)
Voice of Participant
What participants said — dominant theme from each NPS segment.
Promoters
Day 2 content was awesome - much better technically speaking
Passives
Got to see a different side of Claude’s capabilities that I usually see in client delivery
Survey & Data Quality
Survey & Data Quality
Day 1 responses43
Day 2 responses39 (91% of Day 1)
Matched pairs39 (100% of Day 2)
Orgs resolved43 of 43 respondents matched to named org
Organisations
All Day 1 respondents · n=43
IBM8 (19%)
Cognizant7 (16%)
Deloitte6 (14%)
PwC3 (7%)
Salesforce3 (7%)
Forgd.AI3 (7%)
Trace32 (5%)
LTM2 (5%)
AWS1 (2%)
Obin AI1 (2%)
GlobalLogic1 (2%)
McKinsey1 (2%)
Clarinet1 (2%)
DXC Technology1 (2%)
Roberthalf1 (2%)
TCS1 (2%)
Ascendion1 (2%)
Function × Seniority
All Day 1 respondents · n=43
FunctionPractitionerSr PractitionerMgr / Sr MgrDirector+Partner / MDTotal
Engineering523––10
Architecture–747–18
Business Leadership–12238
Project / Engmt––34–7
Experience Profile
AI experience level · n=43 · % of row
By Persona
SegmentExploringApplyingDeliveringFrontiern
Architect—56%28%17%18
Transformation Lead27%53%20%—15
Developer40%30%20%10%10
By Seniority
SegmentExploringApplyingDeliveringFrontiern
Director, Senior Director, or Principal15%69%15%—13
Manager or Senior Manager25%33%25%17%12
Senior practitioner (5–9 years)10%50%20%20%10
Practitioner (0-4 years in role)40%20%40%—5
Partner, Managing Director, or Executive—67%33%—3
How would you describe your current experience with AI tools prior to Basecamp?
Learning and exploring
8 (19%)
Applying in practice
21 (49%)
Delivering independently
10 (23%)
Operating at the frontier
4 (9%)
Prior to today, how much had you worked with Claude or the Anthropic API?
A little
17 (40%)
Regularly
26 (60%)
Did the depth land for this audience?
Technical depth perception · n=43 respondents · % of row
By Persona
SegmentToo basicAbout rightToo advancedn
Architect6%94%—18
Transformation Lead7%80%13%15
Developer10%70%20%10
By AI Experience Level
SegmentToo basicAbout rightToo advancedn
Applying in practice10%90%—21
Delivering independently—90%10%10
Learning and exploring—62%38%8
Operating at the frontier25%75%—4
Overall depth distribution
Too basic
3 (7%)
About right
36 (84%)
Too advanced
4 (9%)
Did the pace work across the room?
Session pace perception · n=43 respondents · % of row
By Persona
SegmentToo slowWell pacedToo fastn
Architect6%94%—18
Transformation Lead—93%7%15
Developer—80%20%10
By AI Experience Level
SegmentToo slowWell pacedToo fastn
Applying in practice—90%10%21
Delivering independently—100%—10
Learning and exploring—88%12%8
Operating at the frontier25%75%—4
Overall pace distribution
Too slow — could have covered more
1 (2%)
Well paced
39 (91%)
Too fast — not enough time to apply
3 (7%)
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=43 · 1–5
Mean end-of-D1 build confidence by Persona
Architect4.61/5 · n=18
Developer4.30/5 · n=10
Transformation Lead4.20/5 · n=15
Programme mean: 3.9/5 · 25 cohorts
Mean end-of-D1 build confidence by AI Experience Level
Operating at the frontier4.50/5 · n=4
Delivering independently4.50/5 · n=10
Applying in practice4.48/5 · n=21
Learning and exploring4.00/5 · n=8
Overall end-of-D1 build confidence distribution
1
0 (0%)
2
1 (2%)
3
2 (5%)
4
19 (44%)
5
21 (49%)
How relevant was today's content to your current role?
Content relevance rating · n=43 respondents · 1 = not relevant, 5 = highly relevant
Mean relevance by Persona
Architect4.72/5 · n=18
Developer4.70/5 · n=10
Transformation Lead4.00/5 · n=15
Mean relevance by AI Experience Level
Applying in practice4.52/5 · n=21
Operating at the frontier4.50/5 · n=4
Delivering independently4.50/5 · n=10
Learning and exploring4.25/5 · n=8
Overall relevance distribution
1
0 (0%)
2
1 (2%)
3
3 (7%)
4
14 (33%)
5
25 (58%)
How likely are you to recommend attending this programme to a colleague?
n=39 Day 2 respondents · 91% of Day 1
Score distribution · 0 = not at all likely · 10 = extremely likely
Cohort012345678910
Cohort 3539522
■ Detractors 0–6■ Passives 7–8■ Promoters 9–10
NPS Breakdown
69%
31%
0%
NPS +69 (n=39)
Promoters (9–10)Passives (7–8)Detractors (0–6)
Programme mean: +42 · 25 cohorts
Promoters mean: 52%
Passives mean: 38%
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)
Cognizant
100%
0%
0%
NPS +100 (n=7)
Deloitte
40%
60%
0%
NPS +40 (n=5)
IBM
29%
71%
0%
NPS +29 (n=7)
By Persona
Promoters (9–10)Passives (7–8)Detractors (0–6)
Architect
75%
25%
0%
NPS +75 (n=16)
Transformation Lead
71%
29%
0%
NPS +71 (n=14)
Developer
56%
44%
0%
NPS +56 (n=9)
Programme means · 25 cohorts: Architect +50 · Developer +50 · Transformation Lead +34
By AI Experience Level
Promoters (9–10)Passives (7–8)Detractors (0–6)
Learning and exploring
86%
14%
0%
NPS +86 (n=7)
Applying in practice
80%
20%
0%
NPS +80 (n=20)
Delivering independently
44%
56%
0%
NPS +44 (n=9)
Programme means · 25 cohorts: Delivering independently +46 · Applying in practice +45 · Learning and exploring +43 · Operating at the frontier +34
What is the main reason for your score?
n=34 responses · organised by NPS segment
Promoters (score 9–10)· 24 responses
Day 2 content was awesome - much better technically speaking
9“Good context on product functionality and capability. I would have loved to have a section on go to market. What’s worked, what’s moved the needle in workplace adoption, what’s created real value. You have a better view of that globally than we would where our focus is more diluted across products and services.”
Passives (score 7–8)· 10 responses
Got to see a different side of Claude’s capabilities that I usually see in client delivery
8“Too short of a time to assimilate and work on the solutions . I think the setup portion and me made a pre-requisite before the session as each org setup may be different”
End-of-Programme Confidence
n=39 respondents · 1 = not confident, 5 = very confident
How confident are you in your ability to design an evaluation for an AI solution?
12345Mean
——8%46%46%4.38
Programme mean: 4.1/5 · 25 cohorts
Leaving today, how comfortable are you advising clients on AI when questions arise?
12345Mean
—3%—51%46%4.41
Programme mean: 4.2/5 · 25 cohorts
How confident are you in your ability to lead a conversation about Anthropic and Claude with a client?
12345Mean
—3%5%41%51%4.41
Programme mean: 4.2/5 · 25 cohorts
Confidence Trajectory by Persona
D1→D2 delta · apply-AI confidence · matched pairs only · n=39
Positive delta = confidence grew · negative = dropped
Architect (n=16)
-0.06
Developer (n=9)
+0.00
Transformation Lead (n=14)
+0.00
What one takeaway will you share with a colleague or client?
n=29 responses
Evals & testing
“Evals!”
Claude / Anthropic content
“Anyone can do it with the help of Claude, and a working api of course”
AI agents & engineering
“Evals and context engineering”
Speed / pacing concerns
“Keep learning and this space is moving fast”
AI strategy & use cases
“Eval, context engineering, choosing right model for different use case to save cost”
What will you build for yourself or apply at work in the next 30 days?
n=30 responses
Evals & testing
“Evals!”
AI agents & engineering
“A personal agent able to oversee my customers”
Claude / Anthropic content
“Claude code with databricks”
AI strategy & use cases
“Definitely going to apply evals and be more purposeful on how I use context understanding that it’s a finite resource”
Tool setup / readiness
“Tools and skills”
NPS Reason
n=34 responses · grouped by NPS segment
Promotersscore 9–10 · 24 responses
1Day 2 content was awesome - much better technically speaking10
2Everything was really good . Understood more on eval and importance of inference tuning .10
3Overall basecamp10
4Learning the latest techniques and validating assumptions is critical to customer engagements.10
5Expedited learning10
6Well organized and great content10
7Lots of new information on evals and context engineering. Found it very useful10
8Great interaction with people and use cases10
9Helped demystify Claude code! Learned a lot about how it works even without much experience and met lots of great people10
10Hands in exercises and in depth understanding10
11It was a wonder to know a lot of concepts covered in just two days10
12This was an incredible training! Learned a lot about ClaudeCode and how to utilize its capabilities!10
13I was using Claude close to 6 months and it opened my mind with more useful features that can reduce and optimize my workload.10
14I found the content very applicable & think that the delivery/instruction was also very good10
15Recommend and benefit client and personally10
16Relevant for all technical/functional levels, walking away with executable next steps for my client projects.10
17Very relevant content. Good balance of hands on work and instruction.10
18Details gathered that I wasn't confident about prior such as Evals10
19It covered a lot of ground with hands on reinforcement exercises.10
20Learned a lot about Claude features in the last 2 days and how to use Claude in the days to day work and how easy it is to develop new ideas with Claude10
21Workshop activities9
22Would like to call out the exercises, especially their emphasis on wringing out efficiency and accuracy across the applied AI lifecycle9
23Great trainers!!9
24Good context on product functionality and capability. I would have loved to have a section on go to market. What’s worked, what’s moved the needle in workplace adoption, what’s created real value. You have a better view of that globally than we would where our focus is more diluted across products and services.9
Passivesscore 7–8 · 10 responses
1Got to see a different side of Claude’s capabilities that I usually see in client delivery8
2It's useful information and context for those building with AI8
3Content8
4Good level of disambiguating. Hands on felt useful. Bouncing ideas off other was helpful to. Really enjoyed the social aspect.8
5Too short of a time to assimilate and work on the solutions . I think the setup portion and me made a pre-requisite before the session as each org setup may be different8
6The workshop provided great detail while also empowering the end users with the frankness of the open ended-ness of where the capabilities stand8
7Some of the concepts go beyond base camp .. this is opening to validate & manage or fine tuning your agents to models8
8Learnt a lot7
9It’s really good networking and the eval track is strong7
10Valuable work examples that are directly applicable in the field7
Most Valuable
What one takeaway will you share with a colleague or client? · n=29 responses
1Evals!
2Do the basecamp
3Evals
4Eval
5Keep learning and this space is moving fast
6Inference optimization pathways
7Use Eval
8Evals and context engineering
9Evals, Inference
10Anyone can do it with the help of Claude, and a working api of course
11Stay to the cutting edge, things evolve fast in this field
12Understanding the different products and the flavors and use the right product to address the relevant problem
132. EVAL and inference optimization
14Almost everything is possible
15How easy it really is to whip up and deploy an agent.
16Don’t blame on models
17How to build better evals
18Eval, context engineering, choosing right model for different use case to save cost
19Make llm outputs as deterministic as possible by following a test driven approach
20Reasoning has gotten WAY better than previously estimated, it’s rarely the model that’s “not good/right” and everything else around it (harness, evals :)).
21Context Engineering g and Agent Evals towards improved AI reliability
22Building evals.
23The importance and value of evals.
24Must attend the session
25No one right way to implement. The lines between some of the levers we can pull (e.g. skills/subagents/mcp) are blurry
26Manage context and Evals are essential for design and successful build.
27Evals
28Its easy to use Claude and learn along the way
29Start trusting the Claude .. be more informed about how to ask claude
30-Day Intentions
What will you build for yourself or apply at work? · n=30 responses
1A personal agent able to oversee my customers
2Evals!
3Claude code with databricks
4TBd
5Tools and skills
6Sales campaign
7An enterprise brain, for our customers
8Automated DQ builder
9Definitely going to apply evals and be more purposeful on how I use context understanding that it’s a finite resource
10Build my own website
11Test my knowledge and continue looking for ways to apply Claude code not just in theory but reality!
12More learning and hands on application of what we learned
13Work more on the new concepts learned
14Evals
15I’ve got a number of internal work projects that I could leverage Claude for to speed up project delivery for clients. And for my own personal agenda, it might be time to purchase a Claude pro max plan.
16Work on context engineering and evals
17I’m working on a data evaluation skill
18Automating operations by using claude
19Eval datasets
20Going to “agentify” the product development process for a health plan client’s flagship program to automate manual adjudication for pended claims.
21Build on the Institutional Knowledge Agent applicable to work
22More experiments on specialist swarms and evals for skills
23I may work on this app we presented in our group project on staffing.
24Agent sales pitch solution for targeting clients based on prior experience with them
25Context manager and eval suite for a project specific research agent.
26I want to build a reusable skill for Claude code, so fellow consultants can benefit from the lessons learned this week.
27I’ll wrap this into existing document aggregation and report generation capabilities
28Makeing the coursework into repeatable skills
29Build apps for the ideas I have
30Try to build an eval.. try and learn all the day2 repo