Alkemy OS

A deal-creation platform for consulting partners

2025
LiveproductsAug 10, 2025Visit site
Next.jsFastAPITanStack QueryPostgreSQLStripe

Alkemy OS is the flagship product at DevDash Labs. Most AI tooling in consulting was built to replace analyst work: automating decks, research, and busywork. Partners never got a tool aimed at the thing they are actually measured on, which is creating deals.

The framing on the product says it plainly: "The analyst stack got AI. The partner stack got nothing."

Alkemy OS is a purpose-built platform to help a consulting partner win deals from their existing relationships. It pulls together account context, current filings, ranked next actions, and the connections that run through a firm's partner network, then puts an account-specific assistant on top that learns from prior conversations and pursuits. Work that used to take an analyst a week comes back as a prepared account packet in a couple of minutes.

What I work on

I'm a core frontend and backend developer on Alkemy OS. My work spans the Next.js application and the FastAPI services behind it. A sample of recent shipping work:

  • Billing and credits. Built out Stripe integration and the credits system that meters usage across the product.
  • Unified signal feed. Replaced per-account checking with a single feed that brings every account's signals together, newest first, with quick actions and brief status inline.
  • Global search. A Linear-style search that spans the whole workspace, so accounts, companies, and past conversations are one keystroke away instead of buried in separate lists.
  • Onboarding. Reworked setup into a single guided screen that auto-populates a workspace from background research, replacing a stack of forms.
  • App UI overhaul. A broad pass on the interface to make dense, research-heavy workflows feel fast and legible.

Approach

The product lives or dies on how much friction it removes for a busy partner, so most of the engineering is about making heavy work feel light: fast search over a large workspace, feeds and briefs that update as data arrives, and a UI that stays readable when there's a lot on screen. The frontend and backend are built as one contract so the interface can stay in step with what the agents and services produce.

Gopal Khadka