Product Growth

Author of 1 newsletter deep dive and 2 podcast episodes a week on how to succeed as a PM, product leader, or aspiring PM.

Lead a team of folks.

I'm the product guy - making sure everything hits our ever increasing quality bar.

Here's a sample of recent content:

Course 1: Mastering the PM Career

Unit 1: The Key Outputs

Product requirements documents
Roadmaps: Advanced techniques
The feature results writeup
The product review bible

Unit 2: Hard Skills

Advanced discovery techniques
How to identify and validate product risk
Advanced customer interview techniques
How to ramp a product team without strategy

Unit 3: Soft Skills

How to say no
Handling conflict as a PM
How to bridge the divide between stakeholders and users

Unit 4: Career Skills

The PM Career Ladder: Your Unofficial Guide
How to get a raise in this market
Why politics isn’t a dirty word

Course 2: Mastering Growth

Unit 1: Lessons From The Greats

How Miro grows
How Canva grows
How Klarna grows
How did Intel do so Badly?
Everything about Midjourney
What you need to know about Temu

Unit 2: Systematizing Our Learnings

Build a detailed growth model
Gaming growth playbook
When to A/B test
Impact Sizing
Growth loops
How to build a flywheel
How to use Code Interpreter

Unit 3: Success in Specific Product Teams

Information Architecture
Product-led sales
Design Systems
Product Copy
Onboarding
Expansion
Activation

Course 3: Product Leadership

Unit 1: How to set up your product team

How to set up your product team
Do you really need a North Star Metric
Pitch a growth role to senior leadership
How founders think differently from PMs

Unit 2: How to scale your team

Billion-Dollar Execution
How to scale a product team from 0 to 15
Reference checks can supercharge your hiring

Unit 3: How to be a great leader

7 currencies of a product leader
The 7 failure patterns of managers
How to Make Your Ideas Stick
Product leadership in India

Newsletter Author, Podcast Host

Apollo.io

Product leader when the company became a unicorn ($750M->$2.5B valuation).

I led several product teams, including the growth team, pricing & packaging, activation, expansion, information architecture, acquisition, and engagement.

Raised at a unicorn valuation while I was there.

Ended up leaving to focus full-time on my newsletter.

VP of Product

Affirm

Member of the senior product leadership team. Led growth team when company revenue grew from $300M to $1.2B/year.

Head of Growth Product

Epic Games

Led data driven growth for Fortnite during some it's best performing years, including the Travis Scott Concert and launch of "Chapter 2."

Lead Product Manager

thredUP

Built and led the team's product growth team when it raised its series D.

We worked on the following key initiatives:

- Pricing optimization: we re-architected the pricing system to optimize for conversion, new customers, and stockthrough levels, driving industry-leading inventory turns
- Conversion rate optimization: we drove 2.5x in first visit to purchase conversion via tools like outfits, dynamic merchandising, ad campaign customization, and the aforementioned pricing system
- Referrals channel: we grew this channel back when it actually worked, with things like $10-$10 and also a famous $0.99 sales
- Paid advertising growth: we built an ad-technology system to handle budgeting and bids cross-channel across Facebook, Pinterest, and Yahoo ads

Overall, I helped the company eventually go public by setting the foundations for its growth, growth marketing, and growth product teams.

Director of Growth

Rap to Beats

Developed iOS app Rap to Beats & scaled to 300K monthly active users (MAU). Led team.

We shipped several game changing features:
- In-app purchases for premier DJ beats
- A leaderboard to showcase your raps by country
- Social sharing to Instagram and Facebook

With premier DJs regularly using the app to sell their work, we were able to scale one of the first consumer apps for rap on the iPhone.

Co-Founder/ Head of Product

Scoutforce Werving & Selectie

ScoutForce was a SaaS for College Athletic Recruiting Teams. 

I joined as one of the first employees, and helped create the initial product: in the cloud recruiting software to enable recruiting teams. I did in depth discovery, wireframing, and iteration on the early features. 

This built my initial skillset in product management.

Product Manager

The Product Growth Guy 🚀 | Helping you land your next job + succeed in your career | Newsletter Writer and Podcast Host

TL;DR: I write a newsletter for product managers, product leaders, and aspiring product managers. I focus on actionable and deep tips to improve your career.

The Longer Version:

After a long career in PM - rising to VP of Product at a Unicorn - I have made the shift to writing about it.

In particular, I write about product growth. That is, the art of growing via your product. 

This means I also talk a lot about product-led growth (PLG), the SaaS B2B motion.

My newsletter doesn't just focus on growth though. 

I also break down technology, product management, product leadership, and how to get a PM job. 

It's all the things that make for a great use of your learning & development budget. 

Or even personal spend. 

The math is pretty great. 
↳ $150 for an annual subscription
↳ $211K average PM salary in the US
↳ 0-10% average range of raises offered

If the newsletter helps you get a 0.07% better raise, it pays itself. 
That's 1/142th of your raise range.

It's a pretty smart investment for every level of PM.

http://www.news.aakashg.com

Aakash Gupta

This is what a Senior AI PM at Google's Github looks like.

It's wild:

Take a look:  https://lnkd.in/e7Jb8kRe

He's made 484 contributions in the last year (>1/day). 

New-age AI PMs don't just write PRDs. They ship:

1. Prototypes
2. Frameworks
3. Perspectives on AI
4. Open source contributions

The bar for technical fluency for AI PMs continues to rise as PMs like Shubham Saboo raise the standard. 

AI PMs don't just have a theoretical understanding of concepts, but a practical knowledge of AI engineering.

Take a look at Shubham's top repos, for inspiration:

1. Collection of LLM Apps: https://lnkd.in/e-Zesb7u
2. MCP Playbook for AI agents: https://lnkd.in/eR5KHvww

These are practical resources that he's collected himself. They aren't necessarily at-scale production apps. They're PM friendly projects. 

He's also built a more-technical open source Python toolkit for building AI agents: https://lnkd.in/eCsAzPtR

When you put these types of work products together, jobs come to you. This is how you break into AI PM. 

If you look at his LinkedIn, most of his career was in dev rel. But a Github speaks louder than titles.

This is what all aspiring AI PMs should do:

1. Start with Prototyping: https://lnkd.in/eJujDhBV
2. Then pick up Cursor: https://lnkd.in/d2pcXD7R
3. Build with Claude Code: https://lnkd.in/eUyPEAma
4. Ship to Github: https://lnkd.in/eVFVT_Yc
5. Create a Portfolio: https://lnkd.in/eT3RvQiM
6. Build a Job Search System: https://lnkd.in/efbppEDj

Then you too can become an AI PM at Google like Shubham. He's created the roadmap. Now follow.

📌 Want my complete guide to the AI PM job search? Comment 'AI job search' + DM Me. Must do both.

➕ Follow Aakash Gupta for daily AI PM tips. 

Proof of work > titles

The AI PM Guy 🚀 | Helping you land your next job + succeed in your career

Linear hit a $1.25B valuation with just 2 PMs.

The way they build product is from the future:

They're contrarian.

They do almost no "growth product" work:

• No A/B tests
• No work on activation
• No growth dashboards
• Zero focus on retention tactics

They also don't have a design system. 

Instead, at the core of the way Linear builds product is "The Linear Method." Many founders I talk to have been religiously copying it. There's three keys, as I see it:

1. Directness
They don't do user stories, they speak clearly

2. No Busy Work
They don't do sprint planning, or OKRs for ICs

3. Momentum
They don't have cycles, scope aggressively, ship fast

Their founders, and CEO Karri Saarinen have operated this way from day 1. They set a very intentional culture. 

Beneath the method, there's tons more texture... 

One of the most unique processes they have is the roast. Before every launch, the entire team spends 30 mins breaking every feature with a massive doc. 

Another one is their zero bugs policy. I heard from multiple people of bugs being solved in hours, not days.

They also don't have a PM on every project. With 2 PMs, small little teams of designers and engineers really think like product people in the work that they do. 

The results speak for themselves.

$1.25B valuation with just 100 employees.
145% net revenue retention. 
No retention team needed.

They've put on a masterclass in building a unique product culture, with lots to learn for others out there.

Want to go deeper?

• My Deep Dive on Linear: https://lnkd.in/e_R9ruDG
• Head of Product Nan Yu: https://lnkd.in/e3WDcETz
• AI Agents Demo: https://lnkd.in/etXnfc2C

📌 Want my breakdown of their PLG iceberg? Comment 'Linear Iceberg' + DM Me and I'll reply to your DM.

I can't help but think most teams will move this direction.

AI + Product Management 🚀 | Helping you land your next job + succeed in your career

Want to pass PM Interviews at OpenAI, xAI, Meta, Google, Airbnb, Dropbox, Palantir?👇 

1. Understand these important concepts:

- Business: P&L, metrics trees
- Technical: APIs, scalability, performance
- Roadmapping: now/next/later, alignment
- AI/ML: prompt eng, RAG, fine-tuning, evals
- Experimentation: when to A/B test, p-values
- Prioritization: RICE, value vs effort, opp cost
- Discovery: customer interviews, user research
- Metrics: north star, guardrail, leading v lagging
- Product strategy: future trends, big tech choices

2. Practice these case interview questions:

- Product sense: https://lnkd.in/eTqEFtPM
- AI product sense: https://lnkd.in/ekWty_bw
- System design: https://lnkd.in/etTmzNzz
- Product design: https://lnkd.in/encHB-k7
- Product execution: https://lnkd.in/e8t8zS9q
- Product estimation: https://lnkd.in/eZvjP4RR
- Product strategy: https://lnkd.in/eqzpahmW
- Success metrics: https://lnkd.in/efdj_QYT
- Vibe coding: https://lnkd.in/e66DrW-h

3. Practice these behavioral questions:

- Tell me about yourself: https://lnkd.in/eYmRYF2Y
- Why want to work here: https://lnkd.in/eeuRsR9M
- What's your current pay: https://lnkd.in/gS-CKH9P
- Improve app speed: https://lnkd.in/dG9EFhiR
- Greatest weakness: https://lnkd.in/ePQ7dBtr
- Were you laid off: https://lnkd.in/e2iTP3xs
- Leadership style: https://lnkd.in/eCPdQVp2
- What level targeting: https://lnkd.in/e2YQ8tK2

4. Study these company specific processes:

- Microsoft:  https://lnkd.in/eXWWYefH
- Amazon: https://lnkd.in/eA_WxXHQ
- Apple: https://lnkd.in/eXWWYefH
- Google: https://lnkd.in/ezAHYXhV
- OpenAI: https://lnkd.in/ebR_Auss
- Meta: https://lnkd.in/evHQ_w-E

Image: Yangshun Tay

———

♻ Repost to help others in your network
📕 Save the post so you can find it in future
📌 Comment 'Interview guide' + DM me for all the links
💡 Follow me Aakash Gupta and my company account Product Growth for 2x daily AI PM insights

🛠️🤖 How to build AI agents from scratch
(Even if you've never done it before.)
𝗧𝗵𝗲𝘀𝗲 𝗮𝗿𝗲 𝘁𝗵𝗲 𝟴 𝘀𝘁𝗲𝗽𝘀 𝘁𝗼 𝘁𝗮𝗸𝗲, 𝗳𝗿𝗼𝗺 𝗽𝘂𝗿𝗽𝗼𝘀𝗲 𝘁𝗼 𝗨𝗜.

𝙁𝙞𝙧𝙨𝙩, 𝙬𝙝𝙖𝙩 𝙞𝙨 𝙖𝙣 𝘼𝙄 𝙖𝙜𝙚𝙣𝙩?

An AI Agent is an LLM that:

• Reasons on its own
• Act (via calling tools)
• Observes & improve itself

This is the core "ReAct" framework Google released all the way back in 2022. It's now a reality. 

---

𝙃𝙚𝙧𝙚'𝙨 𝙝𝙤𝙬 𝙩𝙝𝙞𝙨 𝙩𝙧𝙖𝙣𝙨𝙡𝙖𝙩𝙚𝙨 𝙞𝙣𝙩𝙤 𝟴 𝙠𝙚𝙮 𝙨𝙩𝙚𝙥𝙨:

1. Define purpose & scope

Over-scoping is the killer of agents. You want a single clear use case with success criteria, and constraints. 

2. Build system prompt

This is often the most important step. You want a really good prompt, or the whole thing falls over. 

3. Choose LLM

Don't always go for the best. Balance speed, multi-modality, and what each LLM is best for. 

4. Integrate tools

This is what makes your agent an agent, MCP, APIs, other agents as tools. Create your architecture.

5. Add memory

Don't add memory unless needed. But for support agents that need conversation history, or research agents that build on past findings? Game-changer.

6. Orchestrate things

You want a control flow over routes, workflows, message queues and your agent-to-agent (A2A) coordination.

7. Build a user interface

This is the flashy part we want to jump to. But only it do it now. Figure out how users will interact. 

8. Add testing & evals 

This is the most important step to create a continuous improvement loop. Systematically analyze errors.

Here's guides to help with each step:

1. Purpose ➝‬ https://lnkd.in/eekC6bwT
2. System prompt ➝‬ https://lnkd.in/d_qYCBT7
3. Choose LLM ➝‬ https://lnkd.in/eqRgsMZx
4. Integrate tools ➝‬ https://lnkd.in/ezNsTfbD
5. Add memory ➝‬ https://lnkd.in/eeey5Cxr
6. Orchestrate things ➝‬ https://lnkd.in/eKhmKYNp
7. Build a UI ➝‬ https://lnkd.in/epRC64ku
8. Evals ➝‬ https://lnkd.in/eGbzWMxf

---

𝙏𝙚𝙘𝙝 𝙨𝙩𝙖𝙘𝙠:

Now, what tools should you use? 

For general assistants: ChatGPT, Claude, Perplexity 
For coding projects: Cursor, Windsurf, Claude Code 
For business automation: Lindy, Relay app, n8n 
For complex workflows: LangGraph, CrewAI, LlamaIndex

The best stack is the one that gets you shipping fast.

---

Agents are one of the most powerful skills for PMs and engineers to learn right now. 

I have been studying them deeply.
 These are my core learnings.

📌 Want my 8-layer agent architecture? Comment 'agent architecture' + DM me.

What agents have you been building?

This $4B CPO runs her entire day out of Claude Code + Cursor. And she shared exactly how she does it:

It was a delight to sit down with Rachel Wolan.

She is one of the most AI-forward CPOs I have met, and she dropped all the sauce on her AI workflows.

🎬 Watch Now: https://lnkd.in/es6aQigP

Available Everywhere
Spotify: https://lnkd.in/eyt7agKj
Apple: https://lnkd.in/eAEVwr3u

🏆 Thanks to our sponsor:

Linear: Plan and build product like the best

Use my link! https://lnkd.in/e8rtjVRZ 

✍️ Here were my favorite takeaways:

1. There are two sides to AI product leadership: being a productive AI product leader yourself, and shipping AI native features for your customers. Rachel showed us both in this masterclass.

2. Webflow went from 0% of designers using Cursor to 30% using it weekly after one builder day. That's the power of setting up your org correctly. Champions teach, incentives align, adoption follows.

3. Rachel built an "agentic chief of staff" that analyzes her calendar for delegation opportunities, triages her entire inbox, and queries Snowflake in natural language. She gets answers in seconds without waiting for data scientists.

4. Her calendar agent spots double-bookings, context switching issues, and meetings she should delegate. It recommends what to cut next week. She runs it weekly & shares outputs with her EA.

5. Her email agent removes junk automatically, pins what matters, and drafts replies based on her inbox behavior patterns. It learns what she cares about by watching how she responds.

6. She queries their data warehouse in plain English through Claude Code. "How many sites does this workspace have?" answered instantly. Set up MCP servers for Snowflake & Tableau to enable this.

7. Building custom agents takes minutes. Create a markdown file in your /agents folder, describe what you want, Claude generates it. She built a LinkedIn post generator live on the pod.

8. To enable your team: give them tool access (Cursor, Claude Code), run builder days where everyone demos something new, update career ladders to expect AI usage, and let champions on your team teach others.

9. For shipping AI features, invest heavily in evals. Webflow's code generator broke when they switched models because eval coverage was insufficient. 

10. Choose AI features that leverage your core strengths. Webflow built a code generator using their CMS, design system, and production hosting. That's their differentiation versus every other coding agent.

11. Think distribution first, not last. Answer engine optimization is the new SEO. Feed FAQs to your website, build ChatGPT apps, and make sure product changes get recognized in answer engines from day one.

What's even cooler is she is a paid subscriber to the newsletter! 

Join her: https://lnkd.in/gp2cCv8K

And don't miss the ep for her live demos!

Tag who you think I should have on next 👇

What does a PM building AI features do?

I've been interviewing 2+ weekly for 14 months:

Here's the latest tools, skills, responsibilities, knowledge

𝟭. 𝗥𝗲𝘀𝗽𝗼𝗻𝘀𝗶𝗯𝗶𝗹𝗶𝘁𝗶𝗲𝘀

1a. Discovery + Research
Finding where AI actually adds value takes research

1b. Execution Delivery
Getting AI into production is harder than it looks

1c. Strategy Planning
You're planning the future of AI at your company

1d. Leadership Impact
The best AI PMs shape company strategy

𝟮. 𝗞𝗻𝗼𝘄𝗹𝗲𝗱𝗴𝗲

2a. AI Fundamentals
Understanding what AI can and can't do is critical

2b. Product Patterns
Designing for AI requires different UX patterns

2c. Market GTM
Positioning AI products requires market knowledge

2d. Testing Taxonomy
Testing AI is completely different from testing software

𝟯. 𝗦𝗸𝗶𝗹𝗹𝘀

3a. Context Engineering
Getting AI to understand what you need is half the job

3b. Evaluation Testing
You can't ship without knowing if it actually works

3c. AI Architecture
Choosing the right model & system design matters

3d. Cross-Team Communication
You're the bridge between tech teams & business

𝟰. 𝗧𝗼𝗼𝗹𝘀

4a. Core AI Platforms
Claude, GPT-5, Gemini & NotebookLM are the top tools of the trade

4b. No-Code AI Builders
Zapier, n8n, Lindy, Linear & Replit are the most used

4c. Prototyping/Vibe Coding
Bolt, Cursor, Claude Code & v0 ship prototypes fast

4d. Analytics Monitoring
DataDog, Amplitude & Arize track AI performance

With so many things to do and learn...

PM is really a "do everything" role. It's hard to have work-life balance. Here's resources to stay sane:

𝟭. 𝗥𝗲𝘀𝗽𝗼𝗻𝘀𝗶𝗯𝗶𝗹𝗶𝘁𝗶𝗲𝘀

1a. Discovery → https://lnkd.in/e9QrMEDw
1b. Execution → https://lnkd.in/eajEfG5R
1c. Strategy → https://lnkd.in/egemMhMF
1d. Leadership Impact → https://lnkd.in/eymXeBmE

𝟮. 𝗞𝗻𝗼𝘄𝗹𝗲𝗱𝗴𝗲

2a. Fundamentals → https://lnkd.in/e6zyYugs
2b. Product Patterns → https://lnkd.in/eq8xvCAK
2c. Market GTM → https://lnkd.in/eJj4bx_i
2d. Testing Taxonomy → https://lnkd.in/eKhmKYNp

𝟯. 𝗦𝗸𝗶𝗹𝗹𝘀

3a. Context eng → https://lnkd.in/ebfnDUmi
3b. Prompt eng → https://lnkd.in/d_qYCBT7
3c. Architecture → https://lnkd.in/eeey5Cxr
3d. Communication → https://lnkd.in/eg8anRXa

𝟰. 𝗧𝗼𝗼𝗹𝘀

4a. Claude → https://lnkd.in/eZ6UBbAQ
4b. Prototyping → https://lnkd.in/eJujDhBV
4c. Experimentation → https://lnkd.in/e86mpjGR
4d. No-Code Builders → https://lnkd.in/e_fjvHHN

📌 Want my free 5 day course? Comment EEC + DM me.

Follow Aakash Gupta for daily AI PM tips.

Repost to help others ♻️

Google JUST dropped 70 pgs on context eng and it's 🔥

I summarized it for you:

Stateful, personal AI starts with Context Engineering: assembling the right information at the right time.

There's 7 keys:

—

𝗢𝗡𝗘 - 𝗖𝗼𝗻𝘁𝗲𝘅𝘁 𝗘𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 = 𝗦𝗺𝗮𝗿𝘁 𝗔𝘀𝘀𝗲𝗺𝗯𝗹𝘆

Your LLM context window is prime real estate. Context Engineering optimizes what goes in:

→ User intent + conversation history
→ Retrieved facts (RAG for general knowledge)
→ Long-term memory (user preferences, expertise)
→ Tool outputs and grounding data

The right information, assembled at the right moment.

—

𝗧𝗪𝗢 - 𝗦𝗲𝘀𝘀𝗶𝗼𝗻𝘀 = 𝗬𝗼𝘂𝗿 𝗪𝗼𝗿𝗸𝗯𝗲𝗻𝗰𝗵

A session is a single conversation with:

1. Events (user messages, tool calls, observations)
2. State (conversation history, accumulated context)
3. Lifecycle (create, interact, close)

Think: one task, one session. When you close it, the session ends but memories can persist.

—

𝗧𝗛𝗥𝗘𝗘 - 𝗠𝗲𝗺𝗼𝗿𝘆 = 𝗬𝗼𝘂𝗿 𝗙𝗶𝗹𝗶𝗻𝗴 𝗖𝗮𝗯𝗶𝗻𝗲𝘁

Memory persists across sessions for true continuity:

- Declarative (facts, preferences: "I'm vegan")
- Procedural (how you work: "I debug like Sarah does")

RAG retrieves general facts from the internet. Memory captures YOUR specific expertise and preferences.

Big difference.

—

𝗙𝗢𝗨𝗥 - 𝗠𝗲𝗺𝗼𝗿𝘆 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝗼𝗻 = 𝗘𝗧𝗟 𝗣𝗶𝗽𝗲𝗹𝗶𝗻𝗲

LLMs themselves drive memory creation:

1. Extract relevant info from session conversations
2. Consolidate with existing memories (dedupe, merge)
3. Load into vector stores for retrieval

An LLM-powered ETL pipeline. Automated intelligence extraction from your interactions.

—

𝗙𝗜𝗩𝗘 - 𝗣𝗿𝗼𝘃𝗲𝗻𝗮𝗻𝗰𝗲 = 𝗧𝗿𝘂𝘀𝘁

Every memory needs metadata for production use:

→ Source (which session created this?)
→ Timestamp (how fresh is this?)
→ Confidence (how certain are we?)

Provenance is critical for debugging, trust, and reliability in production systems.

—

𝗦𝗜𝗫 - 𝗥𝗲𝘁𝗿𝗶𝗲𝘃𝗮𝗹 𝗦𝘁𝗿𝗮𝘁𝗲𝗴𝗶𝗲𝘀

Two approaches to memory retrieval:

Proactive: Always include in context (user name, critical preferences)
Reactive: Retrieve on-demand based on query (semantic search, vector similarity)

Smart agents know when to pull vs push. Not everything belongs in every context.

—

𝗦𝗘𝗩𝗘𝗡 - 𝗣𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻 𝗥𝗲𝗮𝗹𝗶𝘁𝘆

Building this at scale requires:

• Privacy: User data isolation, encryption, compliance
• Performance: Caching strategies, batch operations
• Scale: Vector stores for semantic search, not SQL

Google runs this across millions of users. The infrastructure matters.

—

This took me 2 hours.

To go further, you need to learn:

1. RAG vs Finetuning: https://lnkd.in/ebfnDUmi
2. Prompt engineering: https://lnkd.in/d_qYCBT7
3. AI agents: https://lnkd.in/eeey5Cxr

📌 Want me to send you the whole PDF? Comment 'context PDF' + DM me. Repost to cut the line.

Follow Aakash Gupta for daily AI tips.

Context Engineering Masterclass by Google

I literally mapped out the whole agents market. 

These are the top players:

SECTION 1: AI AGENT INFRASTRUCTURE

Development Platforms & Frameworks
• Anthropic, OpenAI, cohere, Stack, Beam, ADEPT, BREV/AN, DeepOpinion, Distyl, E2B, KESTREL AI, Moveworks, Sema4, greptile, lyzr, /dev/agents, itwin, wordware
  
Workflow Automation & Orchestration
• Enterprise automation: Zapier, Airtable, n8n, Relay, Lindy
• Multi-agent systems: fetch, crewAI, Martian, aaru, emergence
  
Agent Infrastructure & Tools
• Web & tool use: composio, Browserbase, exa, tavily
• Data & knowledge: LlamaIndex, vectara, Letta, Across AI, zep
• Authentication: ANON, Keet
• Payments: Skyfire, stripe, paymon, nevermined
  
Evaluation, Observability & Voice
• Eval & monitoring: arize, Langfuse, Patronus AI, COVAL, Agency, Haize Labs, vijil, Vocera, Weights & Biases
• Voice AI: ElevenLabs, Cartesia, Deepgram, Retell AI, VAPI, Bland, Synthflow, Hamming, FIXIE, sarvam, WaveForms

SECTION 2: CORE ENTERPRISE USE CASES

AI Assistants
• OpenAI, Anthropic, Glean, Lindy, HyperWrite, Convergence, DeepSky (Airtable), perplexity, Personal AI, Please
  
Software Development
• Cursor, Cognition, replit, Autify, Magic, lovable, FACTORY, imbue, poolside, Tusk, qodo, Magic Patterns, All Hands, Resolve
  
Customer Service
• Decagon, SIERRA, Voiceflow, COGNIGY, Maven AGI, crescendo, DevRev, parloa, Gradient Labs, yampa
  
Sales & Rev Ops
• 11x, clay, AISDR, ARTISAN, bounti, Agent, DocketAI, flowirks, QUALIFIED, mindy, IOX, tektonic ai
  
Data Analysis
• Talonic AI, Embry, ConnectlyAI, Paradigm
  
People Ops
• Mercor, MERCOR, apriora, Borderless AI, otto, moonhub, Tezi, CONVIN, ToothFairyAI
  
Accounting
• Basis, Hypatos
  
Cybersecurity
• AIRMDR, Bricklayer, Dropzone AI, nullify, prophet, RadiantSecurity, Simbian, torq
  
Marketing 
• ability, firsthand
  
Web Research
• Lutra, Reworkd

SECTION 3: INDUSTRY-SPECIFIC AGENTS

Financial Services
• Concourse, FINNY, Boosted, Brightwave, Greenlite, altiius, Hebbla, Indemn, Powder
  
Legal
• Harvey, &AI, eve, NormAI, Parcha, Spellbook
  
Healthcare
• EliseAI, Anterior, Hippocratic AI, Assort Health, health, Synthpop
  
Supply Chain & Logistics
• DIDERO, FleetWorks, HappyRobot, Naratix, Pixel Robotics
  
E-commerce 
• Naratix, CommerceOS
  
Gaming 
• ALTERA, innovacer, thoughtful
  
Industrials & Infrastructure
• Composable, juno, phaidra, Pixel Robotics

Shoutout to Crunchbase. Used it heavily. 

Want to go further on agents?

1. AI agent building: https://lnkd.in/epRC64ku
2. AI agents for PMs: https://lnkd.in/eeey5Cxr
3. Practical AI agents: https://lnkd.in/eyamsW7a
4. AI agents course: https://lnkd.in/eGw6R5v5
5. Become an AI Agents PM: https://lnkd.in/eekC6bwT
6. Relay app CEO: https://lnkd.in/ed_ji7-W
7. Lindy CEO: https://lnkd.in/eqRgsMZx

Follow Aakash Gupta for more AI

If you want traffic your site in 2025.

Don't use Old SEO.

Millions of queries daily are going to LLMs. 

Weekly users for ChatGPT alone hit 800M+. 

You need to:

1. Get cited by ChatGPT/Claude → GEO
2. Rank in Google AI overviews → AEO
3. Be used by AI agent tools → AIO
4. Convert users upon land → SXO

It's time for companies to embrace the AI search wave.

I'm not saying SEO is dead. In fact, being good at SEO makes you good at all of these things. 

But you need a specific playbook. 

Here's your starting point 👇

🔗 OLD SEO 
↳ We were obsessed with ranking number 1.

S - Search
E - Engine
O - Optimization

The bread and butter here was keywords, programmatic, backlinks, and technical optimization.

The core question behind all work was: "Can we rank higher than competitors?"

It was great for driving high-intent traffic. 

The problems were:

• more traffic without intent
• informational clicks
• algorithm updates

As a result, the funnel looked like:

100 visitors
4% convert
4 signups

💬 NEW SEO
↳ You have to get cited and used by AI platforms. 

You build 4 things:

1. Generative Engine Optimization (GEO)
2. Answer Engine Optimization (AEO)
3. AI Integration Optimization (AIO)
4. Search Experience Optim (SXO)

It's great for durable authority that compounds.

And it can drive:

• Built in trust
• Pre-qualified intent
• Visibility where decisions happen

As a result, the funnel looks like:

50 visitors
24% convert
12 signups

📌 WANT TO MASTER NEW SEO?
↳ I've written a full guide with Josh Grant.

His team at Webflow drive >10% of their signups...
Which convert at 6x the rate via AEO + GEO

Here's the guide: https://lnkd.in/e8X-Qt_C

We cover:

• How to build your team
• How to measure these efforts
• And 3 s-tier tactics you can steal from Webflow

👉 Tag your favorite growth person or SEO agency so they get the goods.

Julius Göllner

Advisor & Investor

REP - Revenue Excellence Partners

Co-Founder

ARRtist

Co-Founder & Managing Director

MOODJA

E-Commerce Fulfillment for fashion and lifestyle brands who strive to offer the best customer experience, starting with the unboxing of their products.

eNugget Ventures

Investment Holding focussing on early stage investments in B2B SaaS.

Founder & Managing Director

F&P Stock Solution GmbH

Member of The Supervisory Board

F&P Stock Solution GmbH is one of the global leaders in whole selling overstocks of shoes, fashion and accessories. Based in Berlin the company was founded in 2013 and is now one of the biggest sellers of overstocks, liquidations, customer returns and stock returns worldwide.

We are directly working with a wide network of brands, retail stores and eCommerce platforms and can offer a wide portfolio of well known and highly demanded brands to our customers. Our philosophy is to build up long lasting and trustworthy relationships with our suppliers and customers, to create and develop effective solutions for overstocks and access inventory. By doing that our highest goal is to protect our partners and brands reputation and have full transparency and control about sales and marketing channels.

To make it as easy as possible for our suppliers and clients, we offer a wide portfolio of additional service such as organization of transports, assistance with customs registration and procedure as well as sorting and quality assurance of goods.

talentsconnect AG

Aufsichtsratsmitglied

Parloa

Cello

comstruct

Gigs

Helping European SaaS Founders to scale in DACH and beyond - Serial Entrepreneur (15x) | Early Stage Investor (40x) | GTM Expert

Besides being a Serial Entrepreneur by heart (16 companies founded, 2 exited) I am trying to empower the SaaS ecosystem in DACH by building the ARRtist Platform (www.ARRtist.net). Furthermore I am an active Angel Investor focusing on helping my ventures to reach PMF and build a scalable Sales Motion in the early stages.

I enjoy to support hands on and am happy to share my knowledge, mistakes and best practices.

Investment Holding focussing on early stage investments in B2B SaaS Solutions as well as D2C Brands.

𝐀𝐮𝐬𝐭𝐚𝐮𝐬𝐜𝐡 𝐠𝐞𝐬𝐮𝐜𝐡𝐭 - 𝐃𝐢𝐧𝐧𝐞𝐫 𝐄𝐢𝐧𝐥𝐚𝐝𝐮𝐧𝐠 𝐟𝐮𝐞𝐫 𝐀𝐠𝐞𝐧𝐭𝐮𝐫𝐢𝐧𝐡𝐚𝐛𝐞𝐫𝐈𝐧𝐧𝐞𝐧 𝐮𝐧𝐝 𝐆𝐅𝐬 𝐢𝐧 𝐁𝐞𝐫𝐥𝐢𝐧

Wir bauen mit REP und einem neuen Projekt in Stealth gerade zwei sehr spannende tech enabled Agencies und ich ziehe aus dem Austausch mit erfahrenen Inhabern wie Robin, Friederike, Tim oder Machiel immer extrem viel Inspiration, Energie und lerne viel.

Ich hätte daher mal große Lust, mit spannenden Agenturinhaber:innen und Geschäftsführer:innen aus Berlin (>10 MA) zusammenzukommen - ganz ohne Agenda, aber mit echtem Interesse an Austausch im Rahmen eines gemütlichen Abends auf Einladung von uns.

Wie denkt ihr über Skalierung im Dienstleistungsmodell?
Welche Rolle spielt AI in eurem Setup?
Was funktioniert im Vertrieb, was nicht mehr?
Wie denkt ihr 2026 über Recruiting?
Was treibt Euch aktuell sonst noch?

Es wäre super spannend, ein paar ehrliche Gespräche in einem Save Space zu führen und sich ohne viel BlaBla auszutauschen.

Wen sollte ich daher unbedingt einladen?
Wer hätte Lust dabei zu sein?

Industrie und Dienstleistung sind erstmal zweitrangig. 

LinkedIn please do you magic - Danke für Eure Hilfe!

Ecosystem Builder, Passionate Entrepreneur, Software Investor

No alternative text description for this image

𝐖𝐚𝐬 𝐦𝐚𝐜𝐡𝐭 𝐦𝐚𝐧, 𝐰𝐞𝐧𝐧 𝐦𝐚𝐧 𝐒𝐭𝐚𝐭𝐢𝐬𝐭𝐚 𝐠𝐞𝐛𝐚𝐮𝐭 𝐡𝐚𝐭 - 𝐮𝐧𝐝 𝐧𝐨𝐜𝐡 𝐞𝐢𝐧𝐦𝐚𝐥 𝐯𝐨𝐧 𝐯𝐨𝐫𝐧 𝐚𝐧𝐟𝐚𝐞𝐧𝐠𝐭?

In der neuen Folge ARRtist on AIR spricht Friedrich Schwandt, Gründer von ECDB, über den Aufbau seines zweiten datengetriebenen Unternehmens, die Gemeinsamkeiten aber auch Unterschiede und nötigen Anpassungen im GTM in 2025.

Sehr beeindruckt hat mich Friedrichs tiefgreifende intrinsische Motivation, nach einem sehr erfolgreichen ersten Venture noch einmal komplett von vorn zu beginnen 

Ein paar spannende insights aus unserem Gespräch: 

𝐏𝐫𝐨𝐛𝐥𝐞𝐦𝐞 𝐢𝐦 𝐌𝐢𝐭𝐭𝐞𝐥𝐩𝐮𝐧𝐤𝐭 𝐬𝐭𝐚𝐭𝐭 𝐮𝐧𝐤𝐨𝐧𝐭𝐫𝐨𝐥𝐥𝐢𝐞𝐫𝐭 𝐅𝐞𝐚𝐭𝐮𝐫𝐞𝐬 𝐛𝐚𝐮𝐞𝐧
Der Startpunkt war kein Interface, sondern ein ungelöstes Marktproblem: fehlende Transparenz über globale Onlinehändler. Wer das richtige Problem wählt, gewinnt automatisch Fokus.

𝐃𝐚𝐭𝐚 𝐅𝐢𝐫𝐬𝐭, 𝐏𝐫𝐨𝐝𝐮𝐜𝐭 𝐒𝐞𝐜𝐨𝐧𝐝
Ein gutes SaaS-Produkt steht auf der Stabilität seiner Daten. Friedrich investierte Monate in Struktur, Qualität und Automatisierung, bevor die erste Zeile Frontend-Code geschrieben wurde.

𝐏𝐫𝐢𝐜𝐢𝐧𝐠 𝐢𝐬𝐭 𝐤𝐞𝐢𝐧 𝐄𝐱𝐜𝐞𝐥-𝐒𝐡𝐞𝐞𝐭
Bei ECDB wird nicht nach Usern, sondern nach Nutzung und Daten-Tiefe bepreist. Das erlaubt Value-Based Sales und macht Wachstum deutlich planbarer.

𝐈𝐧𝐭𝐞𝐫𝐧𝐚𝐭𝐢𝐨𝐧𝐚𝐥𝐢𝐬𝐢𝐞𝐫𝐮𝐧𝐠 𝐟𝐨𝐥𝐠𝐭 𝐝𝐞𝐦 𝐃𝐚𝐭𝐞𝐧-𝐅𝐢𝐭
Neue Märkte werden erst durch lokale Datenerhebung erschlossen, bevor Sales oder Marketing dort investieren. Das minimiert Risiko - und maximiert Relevanz.
	
Viel Spaß beim Reinhören und großen Dank an dich Friedrich. Den Link gibt es wie immer in den comments.

𝐀𝐧 𝐀𝐈 𝐆𝐓𝐌 𝐏𝐥𝐚𝐭𝐟𝐨𝐫𝐦? 𝐘𝐄𝐒!
Big fun to back one of Europe’s most ambitious SaaS teams as well as the Italian SalesTech Superstars ;-)

Crono just launched Crono 2.0, and it is a real step forward for SalesTech. 

An AI GTM platform that allows sales teams to find leads, enrich data, automate multichannel outreach, and close deals faster - now with an AI layer that understands natural language.

200+ sales teams already rely on Crono, and with this launch the company brings enterprise-level power to SMBs and scaleups, combining verified data, smart AI, deep integrations and a product that’s obsessively built around usability.

Huge congratulations to Alex Roggero⚡️, Marco Rosmarini, Lorenzo Tiberi, Marco Maddiona and the whole Crono team - your energy and drive is contagious.

This isn’t just a feature release, it’s a product leap, the next era of SalesTech and great evidence for European design and engeneering power. 

Highly advice to give it a try.

CC: Nick Franklin and Sara Archer @ ARRtist Summit 2025

𝐒𝐞𝐚𝐭-𝐁𝐚𝐬𝐞𝐝 𝐏𝐫𝐢𝐜𝐢𝐧𝐠 𝐢𝐬𝐭 𝐭𝐨𝐭!?

63 % der SaaS-Anbieter in Deutschland rechnen immer noch pro Seat ab.

87 % aber planen, ihr Preismodell in den nächsten zwei Jahren radikal zu verändern.

In der neuen ARRtist on AIR Folge erklärt Dr. Sebastian Voigt von hy, wie sich SaaS-Pricing gerade neu erfindet und was du als Founder jetzt daraus lernen solltest. Zusammen mit OMR Reviews (Kudos an Philipp Schrader, Valeska Wittgen und Marvin Müller) wurden mehrere 1000 Reviews analysiert und über 150 ExpertInnen befragt. 

Was fand ich besonders spannend?

𝐃𝐫𝐞𝐢 𝐏𝐚𝐤𝐞𝐭𝐞 𝐬𝐢𝐧𝐝 𝐰𝐞𝐢𝐭𝐞𝐫𝐡𝐢𝐧 𝐨𝐩𝐭𝐢𝐦𝐚𝐥
Die Daten zeigen klar: Drei Preisstufen (Good-Better-Best) erzeugen maximale Klarheit und fördern Upsells. Mehr als vier überfordert, weniger als drei limitiert dein Wachstum.

𝐙𝐞𝐢𝐠 𝐳𝐮𝐞𝐫𝐬𝐭 𝐝𝐞𝐧 𝐭𝐞𝐮𝐞𝐫𝐬𝐭𝐞𝐧 𝐏𝐫𝐞𝐢𝐬
„Best-Better-Good“ steigert nachweislich die Zahlungsbereitschaft. Der psychologische Anker wirkt stärker als jede Rabattaktion und führt zu 5–10 % höherem Umsatz.

𝐊𝐥𝐞𝐢𝐧𝐞 𝐏𝐫𝐞𝐢𝐬𝐬𝐩𝐫𝐮𝐞𝐧𝐠𝐞 𝐮𝐧𝐭𝐞𝐧, 𝐠𝐫𝐨𝐬𝐬𝐞 𝐨𝐛𝐞𝐧
Kunden sollen leicht ins mittlere Paket einsteigen. 30 % Preisdifferenz zwischen klein und mittel, 100 % zwischen mittel und groß - das verkauft die Mitte am Besten.

𝐒𝐞𝐚𝐭-𝐁𝐚𝐬𝐞𝐝 𝐏𝐫𝐢𝐜𝐢𝐧𝐠 𝐬𝐭𝐢𝐫𝐛𝐭 𝐥𝐚𝐧𝐠𝐬𝐚𝐦
AI und variable Kosten verändern die Spielregeln. Transaktionale Modelle - etwa nach API-Calls, Datennutzung oder Output - ersetzen klassische Lizenzpreise Schritt für Schritt.

𝐑𝐞𝐯𝐞𝐫𝐬𝐞 𝐓𝐫𝐢𝐚𝐥𝐬 𝐬𝐭𝐚𝐭𝐭 𝐅𝐫𝐞𝐞𝐦𝐢𝐮𝐦
30 Tage voller Funktionsumfang, danach Downgrade. Der Verlust (Loss Aversion) erzeugt mehr Conversion als jedes Free-Modell. Wer’s testet, bleibt meist.

Wer Pricing richtig denkt, steigert Umsatz, Retention und festigt seine Marktposition - ohne einen Cent mehr in Marketing zu stecken.

Danke Sebastian für den ehrlichen aber vor allem datengetriebenen Blick hinter die Kulissen moderner Pricing-Modelle. 

Bin gespannt auf Eure Gedanken zum Thema Pricing und eure aktuellen Erfahrungen.

𝐎𝐩𝐟𝐞𝐫 𝐨𝐝𝐞𝐫 𝐎𝐩𝐩𝐨𝐫𝐭𝐮𝐧𝐢𝐭𝐲? 
𝐊𝐈 𝐮𝐧𝐝 𝐌𝐞𝐧𝐬𝐜𝐡𝐥𝐢𝐜𝐡𝐤𝐞𝐢𝐭 𝐢𝐦 𝐃𝐢𝐬𝐤𝐮𝐫𝐬

Gestern Abend hatte ich die Freude, im Rahmen des metafinanz Events „KI, Menschlichkeit und die Zukunft der Arbeit“ aus ARRtist Perspektive über die Chancen und Ängste resultierend aus den aktuellen AI Entwicklungen zu sprechen. 

Während im Gründer Ökosystem der Techoptimismus überwiegt, ist die Unsicherheit in der breiteren Gesellschaft sicher etwas höher.

Los ging es daher mit einem spannenden und energiegeladenen Impulsvortrag von CEO Legende Rainer Göttmann, der aufzeigte wie KI schon heute die Arbeitsweise in der Beratung verändert und wie einige selbst gebaute Agenten bei der metafinanz bereits jetzt mehr und mehr Wertschöpfung und repetitive Arbeiten übernehmen. Lieber Rainer - es war mir eine große Freude Dich kennenzulernen. 

Eine sehr spannende Frage hier: 
Findet man Agenten perspektivisch auch im Organigramm und wenn ja wo?

Im zweiten Teil gab es einen offenen Schlagabtausch im Rahmen eines spannenden und diversen Panels.

Ruth Nuber zeigte auf, welche fundamentalen Ängste bei Mitarbeitenden bestehen, wie man diese überhaupt ans Licht bringt und welche wertvollen Initiativen und Möglichkeiten es gibt, diesen entgegen zu wirken.

Zukunftsforscher Harry Gatterer brachte mich zum Nachdenken mit seiner These - das wir als Gesellschaft schon jetzt in einer Art Opferrolle in Bezug auf KI stehen und wir hier schnellstmöglich das Heft wieder selbst in unsere Hand bringen müssen. Nicht die KI treibt uns. Sondern wir betreiben die KI.

Lieber Martin Tomaszewski - vielen, vielen Dank für die tolle Zeit in Wien, die vielen inspirierenden Gespräche und die tollen Menschen, die ihr da gestern zusammengebracht habt. Super gelungenes Konzept!

Opferrolle oder Opportunität - was überwiegt bei Euch?

𝐉𝐞𝐭𝐳𝐭 𝐋𝐢𝐯𝐞: 𝐔𝐧𝐬𝐞𝐫𝐞 𝐇𝐨𝐭 𝐕𝐨𝐢𝐜𝐞𝐬 𝐳𝐮𝐦 𝐂𝐨𝐥𝐝 𝐂𝐚𝐥𝐥𝐢𝐧𝐠

Telefonakquise ist aufgrund zunehmend gesättigter schriftlicher Känäle wichtiger denn je. Das haben wir auch auf dem Sales Summit in Hamburg vor 2 Wochen gemerkt.

Kein Thema wurde mehr diskutiert.
Bei keinem Thema mehr nach Lösungen gesucht.

Grund genug für uns ein kleines Knowledge Hub rund um das Thema Cold calling ins Leben zu rufen. 

Ab dieser Woche geben die führenden Köpfe in DACH ganz praktische Tipps zum Thema Cold Calling - einfach verständlich und direkt anwendbar.

Los geht’s mit einem spannenden Artikel von Anwar El Younoussi zum Thema „Winner Mindset“.

Klar ist: 
Vertrieb ist kein Spiel für Schönwetterverkäufer.
Absagen, Einwände, Funkstille - all das gehört dazu. 
Aber der Unterschied entsteht nicht im Pitch. Sondern im Kopf.

Anwar teilt, was oft viel zu kurz kommt:

✅ Warum Resilienz wichtiger ist als jede Technik
✅ Warum Rückschläge kein Scheitern, sondern Feedback sind
✅ Und wie klare Vorbereitung Sicherheit schafft - gerade in schwierigen Situationen

Ob du gerade erst startest oder schon 100.000 Calls gemacht hast: Mindset schlägt Methode. Jeden Tag. In jedem Call. Den Link findet ihr in den Comments.

Ich freue mich auf euer Feedback und aktuellen Challanges. Wen sollten wir darüber hinaus unbedingt als Hot Voice interviewen?

taplio

The Battle

Aakash Gupta

Julius Göllner

Followers

Posts

Engagements

Likes

Performance ScoreAakash Gupta's LinkedIn Performance

Performance ScoreJulius Göllner's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Aakash Gupta

Julius Göllner

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Aakash Gupta's LinkedIn Performance

Performance ScoreJulius Göllner's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreAakash Gupta's LinkedIn Performance