Skip to content

Technology

Conversation is the interface.
Math is the product.

Large models are gifted talkers and unreliable astronomers. We split the stack hard: the Astro Engine computes on a Swiss Ephemeris foundation; the intelligence explains. Warmth without vagueness. Speed without costume.

See the pipeline Enterprise access
Core
Swiss Ephemeris
Surface
3,000+ APIs
Class
~50ms · edge
Policy
Owned, not rented

The hard split

An LLM can sound wise while inventing a planetary degree.
That is unacceptable.

For a company whose product is guidance under real pressure — a lease, a launch, a resignation — the sky must be cast, not improvised. So we never ask a language model to be an ephemeris.

Conversation surface connected to a computational engine core

Layer A · Language

Explain

Memory, tone, continuity, plain-language windows. The part that sits with you on WhatsApp and the web — without inventing longitudes.

Layer B · Engine

Compute

Positions, houses, dashas, transits, muhurat — proprietary cast on Swiss Ephemeris. The part that makes the answer true.

01

Measured

If we claim precision, the numbers stay public.

Founder-verified production claims. Dated. Not round-number folklore. Same engine that answers on WhatsApp.

3,000+

REST API endpoints

The computational surface of the Astro Engine — proprietary, in production.

Verified

~50ms

Global response class

Chart compute at the edge — Cloudflare, 300+ locations worldwide.

Verified

8

Ayanamsa systems

Lahiri, Raman, KP, Yukteshwar, Fagan-Bradley, Galactic Center, True Chitra and more.

Verified

99.9%

Uptime target

Production reliability for the engine that powers every reading.

Verified

Also in production: 9+ disciplines · 100+ report types · multi-ayanamsa cast · edge delivery (300+ locations).

02

Request path

From a human question to a calendar window.

Four beats. Same spine whether the face is WhatsApp, web, or enterprise API.

  1. 01

    Ingress

    Birth data or a saved chart context arrives from WhatsApp, web, or API. No sun-sign shortcut.

    ctx.load(chart) · lagna

  2. 02

    Cast

    The Astro Engine computes positions, houses, and divisionals on a Swiss Ephemeris foundation — owned, not rented.

    cast(D1…D60) · ayanamsa

  3. 03

    Reason

    Dashas, yogas, doshas, transits, muhurat windows — the stack that answers “when,” not only “what.”

    house · transit · dasha

  4. 04

    Explain

    The intelligence layer turns computation into calm language. Warmth without inventing degrees.

    reply · window · memory

Foundation

Swiss Ephemeris at the core.
Our stack around it.

We do not wrap a third-party “astrology API” and call it proprietary. The Astro Engine is the computational brain of everything we ship — Python/Flask service surface, ephemeris-grade casting, multi-system coverage, edge-aware delivery.

Precision mechanism dissolving into light filaments — ephemeris metaphor
03

Architecture decisions

Four choices we will not reverse.

  • Own the ephemeris path

    Chart casting, houses, dashas, and transits run on computation we control. Product quality is not gated by a vendor roadmap.

  • Multi-system by design

    Vedic, KP, Nadi, Lal Kitab, numerology, Vastu, Muhurat — first-class citizens of one engine, not plugins bolted on later.

  • Edge-aware conversation

    Latency class built for dialogue. A WhatsApp answer should feel like a reply, not an overnight batch job.

  • One brain, many faces

    Astro Ratan, Grahvani, reports, and enterprise APIs share the same math. No marketing engine vs real engine.

Delivery

Conversation speed
is a feature.

Chart compute is served in a ~50ms class from the edge — hundreds of locations worldwide — so a reading feels like dialogue. WhatsApp is a first-class product surface, not an afterthought bolted to a slow core.

Response class
~50ms
Edge footprint
300+ locations
Uptime target
99.9%
Global edge network arcs — latency as product craft

Coverage

What the engine actually speaks.

Production surface — not a marketing checklist of systems we “plan to support someday.”

  • Natal & divisional charts (D1–D60)
  • 5-level Vimshottari dasha
  • Yogas & doshas (classical combinations)
  • 8 ayanamsa systems
  • KP, numerology, Vastu, muhurat
  • Ashtakavarga, Special Lagnas, Sudarshan Chakra
  • Transit & horary
  • Edge cases: polar latitudes, midnight births, date line, retrogrades

Ayanamsa systems

Eight systems in production so practitioners and products can choose the convention they already trust — without forking the engine.

  • Lahiri
  • Raman
  • KP
  • Yukteshwar
  • B.V. Raman
  • Fagan-Bradley
  • Galactic Center
  • True Chitra
04

Faces of the same brain

Where this stack shows up.

  • Consumer · live product

    Astro Ratan

    Live

    Already launched. Live readings, 5:30 AM briefings, continuity — WhatsApp and astroratan.ai on this engine.

    Try live on WhatsApp →

  • Professional

    Grahvani

    Coming soon

    Practitioner depth, reports, academy — same cast, different desk.

    See ecosystem →

  • Builders

    Enterprise APIs

    License the surface. Exact contracts with access — no public sandbox theater.

    Developers →

Next step

Need the sky as infrastructure?

Enterprise access to cast, transit, dasha, muhurat, and report endpoints — the same surface under Astro Ratan. Write with use case, volume, and regions.

Developers Email info@astrocorp.in Watch a reading on home