Ryniant Raises $300K to Rebuild
Synthesis Planning from the Route Up

Founded by two 17 year olds and a 20 year old, Ryniant has raised $300K in pre-seed funding to build the operational layer that sits on top of retrosynthesis. The round was led by Verdict Capital (Niko Bonatsos, Michael Fertik), with Ryniant selected into the Palantir Startup Fellowship.

This capital funds the build and launch of a synthesis planning platform that does not stop at generating a route. It optimizes conditions on every step, checks equipment feasibility against the lab that will actually run the chemistry, adjusts costs for what is already on the shelf, validates every candidate through 14 chemical checks, and exports a protocol ready for the bench. The product ships this summer.

The Problem We Are Solving

Every retrosynthesis tool on the market does the same thing: takes a target molecule and proposes a route back to commercially available starting materials. Some do it with expert rules. Some with template-based search. Some with neural networks. The route is the part everyone has built.

The route is also the part that is not the bottleneck.

The bottleneck is everything that happens after the route appears on screen. A chemist at a CRO receives a route and then spends hours adapting it: checking whether their lab has the right equipment for each step, pricing out reagents at their actual purchase quantities, looking at which solvents and catalysts are already on the shelf, evaluating whether a weak step has a better documented alternative, writing the protocol, running safety checks. One molecule. One chemist. One full day of planning before any lab work begins.

In a drug discovery program with hundreds of candidates, that adds up to months of expert time spent on logistics, not chemistry.

No existing tool addresses this. Reaxys and SciFinder are literature databases with retrosynthesis bolted on. Spaya generates routes and shows some pricing. ASKCOS and SYNTHIA generate routes. They all stop at the route. The chemist still does everything else manually.

What Ryniant Builds

Ryniant is the operational layer for synthesis planning. A chemist inputs a target molecule. The platform outputs ranked routes where every step has been individually optimized, validated, and costed for that specific CRO's lab.

Route generation

A retrosynthesis engine evaluates thousands of disconnection paths simultaneously through tree search with backtracking. It finds structurally optimal routes to commercially available starting materials. This is the foundation, not the product.

Conditions optimization on every step

For each step in the route, the system searches millions of documented reactions for precedents where a structurally similar transformation was run under known conditions. It pulls real yields, solvents, temperatures, catalysts, and reaction times. If no documented precedent exists, an AI chemist proposes conditions and validates its own suggestions through multiple rounds of chemical reasoning. Every candidate is compared against the original on four metrics: predicted yield, reagent cost, equipment feasibility, and compatibility with the next step in the sequence.

Alternative reaction discovery

For each step, the platform asks a harder question: is there an entirely different reaction that produces the same intermediate more efficiently? Not different conditions on the same chemistry. A completely different disconnection, different starting materials, different reaction type. The intermediate is fixed, so everything downstream stays untouched. When a better alternative exists, it is shown alongside the original. The chemist chooses.

14-point chemical validation

Every proposed condition set passes through 14 checks before the chemist sees it. Product verification against the expected structure. Catalyst compatibility with functional groups elsewhere in the substrate. Solvent boiling point versus proposed temperature adjusted for the CRO's equipment. Atmosphere and reagent handling requirements. Stoichiometry sanity. Scale compatibility. Downstream compatibility with the next step. Protecting group consistency across the full route. Full sequential chain verification.

Nothing reaches the chemist without passing all applicable checks.

CRO-specific operations

• •Equipment feasibility. Every step is checked against the CRO's actual hardware profile. A reaction that requires a glovebox or a pressurized reactor is flagged if the lab does not have one. Flagged, not silently dropped, because the chemist may have access through a partner facility.
• •Inventory integration. The CRO uploads their chemical inventory. Reagents already on the shelf are costed at zero. The route cost reflects what they actually need to buy, not the textbook number.
• •Safety assessment. GHS classifications, reaction hazard profiles, and handling requirements for every step.
• •Protocol export. A downloadable lab protocol with exact quantities, temperatures, reaction times, solvent volumes, and safety warnings. Ready for the bench.

The differentiator is not the route. Every competitor generates routes. The differentiator is everything that happens between the route and the bench: optimization, validation, costing, equipment checks, and the protocol. That operational layer does not exist anywhere else.

Why This Matters for CROs Specifically

Contract research organizations synthesize molecules for clients. Their margin depends on two things: how fast they can plan a synthesis and how efficiently they can execute it. The planning step is where Ryniant sits.

A CRO chemist using Ryniant does not receive a generic route from the literature. They receive a route that has been optimized for their specific lab: their equipment, their inventory, their supplier pricing, their scale. The protocol is ready to hand to a technician. The cost estimate is ready to send to a client.

That is the difference between a research tool and an operations tool. We are building the operations tool.

The Data Layer

Conditions optimization is only as good as the data behind it. The platform currently runs on millions of reactions from open and licensed databases. Each reaction is indexed by reaction type and searchable by structural similarity in milliseconds through a pre-built approximate nearest neighbor index. When Ryniant searches for precedent conditions for a given step, it is searching a curated, partitioned, indexed corpus of real experiments, not generating text.

The data coverage expands as we sign additional licensing agreements. The architecture is designed so that every new data source makes every existing feature better without changing any code. More reactions means more precedents, better similarity matches, more accurate yield predictions, and fewer fallbacks to AI-generated conditions.

The Team

Ryniant was founded by three co-founders who have been building at the intersection of software and life sciences since high school. We are advised by medicinal chemists with direct CRO operational experience and supported by academic advisors in computational chemistry and bioprinting.

We are a Palantir Startup Fellowship company, building on Palantir's Foundry and AIP platforms for data orchestration and agentic workflows.

What Comes Next

1. 01Ship the product this summer. Real routes, real data, real protocols, running on real CRO equipment profiles.
2. 02Onboard the first paying CRO customers across India, the US, and the UK.
3. 03Expand the reaction database through additional data licensing partnerships.
4. 04Build the conditions optimization engine to v1: fingerprint-based similarity search with validated fallback for novel substrates.
5. 05Demonstrate measurable time savings in head-to-head comparisons against manual planning workflows.

Work With Ryniant

If you run a CRO synthesis team and your chemists spend hours planning routes that a machine should optimize, we should talk. If you lead a drug discovery program and want to know the real manufacturing cost of every candidate before committing to a single reaction, we should talk.

For pilot partnerships, the fastest path is naitikmohanty@ryniant.com.