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.

BrightMind Enrichment

Architected and deployed 12+ advanced dashboards in SAP Analytics Cloud (SAC) integrated with BW/4HANA live connections, delivering real-time insights on student KPIs and boosting program completion rates by 22%. Built driver-based SAC Planning models leveraging Native HANA calculation views and HANA Cloud Smart Data Access (SDA), improving semester forecasting accuracy by 30% across 500+ enrollment profiles. Orchestrated data pipelines using dbt, Apache Airflow, and Data Intelligence Cloud to automate ETL from BW/4HANA to SAC, reducing manual refresh cycles by 40% and enhancing reporting agility for 3 academic teams. Configured Advanced Data Actions, Multi-Actions, and Predictive Scenarios within SAC to model tuition revenue and resource allocation, enabling the finance unit to adjust budgets dynamically across 15 programs. Engineered custom HANA Calculation Views and optimized Native HANA SQL Script for SAC Planning, accelerating query performance by 35% on high-volume student lifecycle datasets. 
• Piloted requirements workshops with faculty and IT, mapping BW/4HANA Info Providers, Composite Providers, and open ODS Views to SAC stories, standardizing key metrics across 12 curriculum modules. 
• Organized self-service analytics enablement by conducting SAC Story Builder and Smart Predict training for 25+ instructors, driving a 45% uptick in adoption of on-demand insights and data-driven interventions. Partnered with developers to integrate Live Data Connections between BW/4HANA and SAC using Cloud Connector and HANA Smart Data Integration (SDI), improving data latency by 25% for live operational dashboards.

SAP SAC Consultant

GI Junk Removal

Innovated cost efficiency reports in SAP SAC connected to BW/4HANA layers, analysing over 200 service routes weekly and helping reduce fleet fuel expenses by 18% within two months. Modeled disposal fee structures using Native HANA views, streamlining vendor payment tracking for five waste categories and cutting monthly landfill cost overruns by approximately 12%. Integrated historical pickup volumes into BW/4HANA Composite Providers, feeding SAC scorecards that improved daily scheduling accuracy for 150+ service requests across multiple city zones. Instituted dynamic allocation rules in SAC with HANA calculation views, supporting scenario comparisons that identified $3,000 in hidden cost savings through optimized recycling partnerships. Collaborated with operations managers to automate expense audits by linking route data from BW/4HANA to interactive SAC dashboards, shortening monthly reporting cycles by 35% for finance teams.

Business Analyst

MBA Projects

Moderated a market entry feasibility project for a Boston-based fintech client, analysing 3 competitor landscapes, 5 customer segments, and delivering a 30-page report with strategic recommendations to boost market share by. Led a cross-functional team of 4 MBA peers on an operations optimization project, mapping 7 key workflows for a local manufacturing client, reducing process inefficiencies by 22% through lean methodology recommendations. Innovated and executed a pricing strategy project for a SaaS startup, performing detailed break-even and sensitivity analysis on 4 pricing models, increasing projected revenue streams by 18% within 6 months. Developed a 5-year financial forecasting model for a healthcare startup project, integrating data from 6 departments and presenting a 20-slide deck to stakeholders that supported a successful $2M funding round. Delivered a growth strategy project for an edtech firm targeting Gen Z learners, conducting 120 customer surveys and synthesizing insights into a 40-page actionable playbook that improved user acquisition by 25%.

MBA Consultant

Cognizant

Facilitated over 50 stakeholder workshops and requirement-gathering sessions, aligning cross-functional teams and ensuring 98% adherence to project milestones within dynamic Agile delivery environments. 
• Conducted data-driven impact assessments and root cause analysis for process inefficiencies, recommending solutions that reduced operational costs by 18% for a top-tier banking client. 
• Established and executed 120+ user acceptance test cases, coordinating with QA teams to resolve defects, which increased overall system stability by 30% during multiple release cycles. Created interactive dashboards and performance reports using advanced Excel and Power BI, enabling business 
leaders to monitor KPIs and make informed decisions, driving a 22% boost in client satisfaction scores. 
• Systematized repetitive reporting tasks with VBA and SQL queries, saving 200+ hours annually and streamlining month-end reconciliation for high-volume financial transactions. Mentored a team of 5 junior analysts, conducting weekly knowledge-sharing sessions and ensuring quality deliverables, which strengthened team productivity by 40% across complex multi-phase projects. Supported migration of legacy systems to cloud-based solutions, collaborating with developers and solution architects to integrate 15+ modules, resulting in a 35% reduction in system downtime.

OpenText

Assisted in developing and testing enterprise content management modules for OpenText’s Documentum platform, collaborating with a 6-member team to enhance system performance and reduce data retrieval time by 15%. Designed and implemented automated test cases covering 80% of functional scenarios, utilizing Java and Selenium, which helped identify and resolve over 120 software defects during the internship period. Backed senior engineers in deploying secure cloud storage solutions for over 10 enterprise clients, applying SQL and XML to manage structured data efficiently and ensure 99% data integrity. Analysed and optimized backend scripts using Java and Shell scripting, contributing to a 20% improvement in batch processing speed for large-scale document archiving tasks. 
• Prepared detailed technical documentation and weekly reports for 4 ongoing projects, enhancing cross-team communication and accelerating project delivery timelines by approximately 10%.

Engineering Intern

I am a Business Analyst and SAP SAC Consultant with over 7 years of experience helping companies make better decisions through data. I have worked in education, consulting, waste management, and IT services to improve business performance using smart analytics and reporting tools.

My key skills include building dashboards in SAP Analytics Cloud (SAC), automating data pipelines, and working with HANA and BW/4HANA systems. I’ve created reports and planning models that helped teams save time, reduce costs, and make faster decisions.

I recently completed my MBA at Babson College, where I focused on business analytics and machine learning. During my projects, I helped real companies with pricing strategies, process improvements, and financial forecasting.

Before my MBA, I worked at Cognizant as a Business Analyst, leading requirement sessions, creating reports, and supporting cloud migration for global clients. I also enjoy training teams, improving workflows, and building strong relationships across departments.

taplio

The Battle

Aakash Gupta

Akshay Kalyankar

Followers

Posts

Engagements

Likes

Performance ScoreAakash Gupta's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Aakash Gupta

Akshay Kalyankar

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

Ready to grow your LinkedIn presence?

Performance ScoreAakash Gupta's LinkedIn Performance