AI PM: Ship Your First Customer-Facing Agent

Bharat Batra

Product Leader at Superhuman AI

Akhil Raj Gupta

Founder at Luminary (ex Airbnb)

Anyone can build a prototype. Paying customers are a different beast.

Founding Cohort Perks - 1:1 sessions, attractive price (next cohort at $1K USD), and lifetime access to future runs

Anyone serious about AI in 2026 can build prototypes for internal users.

Putting one in front of paying customers at scale is a different beast altogether: the failure modes multiply, the polish bar is much higher, and nobody is watching over the model's shoulder when things go wrong.

This course helps you take that leap. Together we build a real customer-ready agent — scoped, polished, evaluated, and instrumented — using a working codebase crafted by Luminary instructors.

By demo day, you'll know how to elevate your product beyond "just a wrapper", and how to launch and iterate without burning trust — with guardrails, evals, and observability built in from day one.

The build, week by week:

Week 1 — Make It Work: your agent v0, running

Week 2 — Make It Yours: proprietary context, UX, and tools

Week 3 — Make It Reliable: your eval set + a failure-mode playbook

Week 4 — Make It Safe: guardrails, observability, and a safe launch

What you’ll learn

Move beyond internal demos and ship agents that survive contact with real customers.

Your agent v0 running by week one — then scoped, polished, and made proprietary.
Build on a working codebase crafted by Luminary instructors.
Demo day: present your agent to a panel of experts.

An eval set that reflects real user behavior — not just happy paths.
A failure-mode playbook: prompt, tools, context, self-healing.
The craft that elevates your product beyond 'just a wrapper'.

Observability: see what your agent is doing before users tell you.
Guardrails and safe rollout to contain the blast radius.
Leave with your day-zero launch checklist.

Learn directly from Bharat & Akhil

Bharat Batra

Contact

Product Leader at Superhuman AI

Akhil Raj Gupta

Contact

Founder at Luminary, ex Airbnb

Previously at

See all products from Akhil

Who this course is for

AI PMs & Builders — shape the product, not just the roadmap, and develop the taste to know what's worth building.
Engineers & Tech Leads — master the architecture and eval frameworks that separate real AI teams from vibe coders.
Founders & Operators — go beyond the wrapper: scope, ship, and own the workflow that makes your product hard to replace.

What's included

Live sessions

Learn directly from Bharat Batra & Akhil Raj Gupta in a real-time, interactive format.

Lifetime access

Go back to course content and recordings whenever you need to.

Community of peers

Stay accountable and share insights with like-minded professionals.

Certificate of completion

Share your new skills with your employer or on LinkedIn.

Maven Guarantee

Your purchase is backed by the Maven Guarantee.

Course syllabus

7 live sessions • 12 lessons

Week 1

Sep 10—Sep 13

Sep

Live Lesson with Bharat & Akhil : Make It Work

Fri 9/113:30 AM—5:00 AM (UTC)

Make It Work

3 items

Week 2

Sep 14—Sep 20

Sep

Optional: Office Hours

Tue 9/153:30 AM—4:30 AM (UTC)

Optional

Sep

Live Lesson with Bharat & Akhil: Make It Yours

Fri 9/183:30 AM—5:00 AM (UTC)

Make It Yours

3 items

Schedule

Live sessions

2-4 hrs / week

Four live sessions with Bharat & Akhil, Sep 11 to Oct 2, 90 minutes each, all recorded. Weekly office hours cover you between sessions.

Fri, Sep 11
3:30 AM—5:00 AM (UTC)
Tue, Sep 15
3:30 AM—4:30 AM (UTC)
Fri, Sep 18
3:30 AM—5:00 AM (UTC)

Projects

2-4 hrs / week

A guided build each week — your agent compounds from v0 to a fully instrumented product.

The build, week by week

Week 1 — Make It Work (Fri Sep 11): Skip the theory — start inside a working customer-facing agent, taking it apart piece by piece. The anatomy: core loop, tool calls, context, memory, system prompt. The build-or-plug call: when your product needs its own agent, and when an MCP inside a general assistant is enough. You leave with your agent v0 running.

Week 2 — Make It Yours (Fri Sep 18): Turn your agent into a harness that beats generic Claude Code / Codex in your environment. Design the core loops only you can own, and cut everything else. Bake in proprietary context, UX, and tools a horizontal assistant can't match.

Week 3 — Make It Reliable (Fri Sep 25): Trust is built before launch. Build an eval set that reflects real user behavior, not just happy paths. Diagnose failure modes and pick the right fix — prompt, tools, context, self-healing. You leave with your eval set + a failure-mode playbook.

Week 4 — Make It Safe (Fri Oct 2): Production is where agents earn their keep or burn their brand. Instrument observability so you see issues before users tell you. Set guardrails and roll out safely. You leave with a fully instrumented agent + your day-zero launch checklist.