MATERIALS DISCOVERY + ENGINEERING INTELLIGENCE

Discover what's next.
Understand what makes it work.

Obelisk is an AI-assisted materials discovery and engineering intelligence platform focused on high-temperature aerospace. Explore candidate materials and connect chemistry, structure, synthesis, processing, and degradation risks in one technical workflow.

Proposed candidates. Inspectable reasoning. Laboratory validation still required.

FROM REQUIREMENT TO TESTABLE DIRECTION
01
Define the operating envelopeTemperature, stress, atmosphere, lifetime, and processing constraints.
02
Explore candidate spaceProposed compositions and available structural models.
03
Screen feasibilityChemistry, geometry, toxicity concerns, and potential incompatibilities.
04
Connect processing to performanceSynthesis routes, heat treatment, phases, and degradation mechanisms.
05
Compare and redesignPerformance, manufacturability, cost, risk, and proposed alternatives.
06
Build the validation handoffEvidence, limitations, prioritized tests, and an engineering dossier.
WORKFLOW OVERVIEW · NOT A LIVE SOLVER EXECUTION
Requirement-driven evaluationProcessing-aware analysisExplicit uncertaintyHuman validation required

CAPABILITIES / SEVEN CONNECTED DISCIPLINES

Not just a composition.
The engineering behind it.

From candidate discovery to engineering decisions. Explore how a material could be made, where it could fail, and what to investigate next.

01

Materials Discovery

Explore proposed compositions and structures against high-temperature aerospace requirements.

02

Manufacturing Intelligence

Examine synthesis routes, processing constraints, and heat-treatment strategies.

03

Degradation Analysis

Investigate creep, oxidation, fatigue, and potential failure mechanisms.

04

Validation Planning

Identify relevant calculations, characterization methods, and laboratory tests. Planning supports testing; it does not replace it.

05

Candidate Redesign

Propose lower-toxicity alternatives or changes for a demanding application, while exposing new tradeoffs. Redesign is not a safety certification.

06

Tradeoff Intelligence

Compare performance, manufacturability, cost, and risk to prioritize promising directions.

07

Structure Exploration

Inspect available atomic coordinates and clearly labeled structural models through interactive visuals. Conceptual models are distinguished from validated structures.

REDESIGN / EXAMPLE REQUEST

“Propose a lower-toxicity version of this candidate. Explain what changes, what performance could be lost, and what needs testing.”

TWO WAYS TO WORK

Explore a new candidate.
Interrogate an existing material.

01 / DISCOVERY

Start with the requirement.

Bring a temperature, atmosphere, processing constraint, or performance target. Explore proposed material directions and the tradeoffs behind them.

02 / ENGINEERING INTELLIGENCE

Start with the material.

Bring a known or proposed material. Examine manufacturing and synthesis, degradation risks, redesign options, and a validation plan.

Available analyses depend on the input, supporting data, and configured tools. A proposed direction is not a proven material.

01 / THE BOTTLENECK

Every physical iteration costs time.
Make it count.

High-temperature aerospace development requires costly synthesis, heat treatment, microscopy, mechanical testing, and long-duration exposure studies. Teams must connect these experiments with scattered literature, simulation results, and processing knowledge. Investigating an unsuitable candidate can consume valuable equipment time and budget.

Obelisk brings those decisions into one reviewable workflow: explore candidates, surface potential failure mechanisms, compare manufacturing options, and prioritize what deserves further calculation or laboratory testing.

THE CHALLENGE

Separate data sources → manual reconciliation → repeated screening → costly experimental iterations

THE OBELISK WORKFLOW
  1. Engineering requirements
  2. Composition and structure exploration
  3. Chemistry and geometry checks
  4. Synthesis and processing assessment
  5. Degradation screening
  6. Tradeoff ranking
  7. Candidate redesign
  8. Evidence and uncertainty review
  9. Prioritized validation plan
  10. Engineering dossier

Analysis depth depends on available data and configured tools. Proposed candidates still require validation.

02 / METHODOLOGY

One decision chain.
Six controlled stages.

Obelisk is designed to preserve the connection between what a material must do, how it could be made, why it could fail, and what must be tested next.

Explore the six-stage workflow +
01

Define requirements

Temperature, stress, atmosphere, lifecycle, cost, and processing limits become explicit constraints.

02

Generate candidates

Proposed compositions and structures enter a controlled candidate workspace.

03

Screen feasibility

Chemistry, geometry, toxicity, and obvious stability conflicts are surfaced early.

04

Evaluate behavior

Processing, phase stability, creep, oxidation, fatigue, and degradation are examined together.

05

Rank tradeoffs

Performance, manufacturability, confidence, and risk are compared without hiding conflicts.

06

Produce a dossier

Assumptions, evidence, uncertainty, and required validation tests become one engineering record.

03 / TECHNICAL OUTPUT

See the output.
Inspect the limits.

A result is useful only when the composition, structural model, manufacturing context, failure mechanisms, and validation requirements remain inspectable.

  • Clearly labeled conceptual structure
  • Processing and phase context
  • Guardrails and validation tests
  • Exportable technical artifacts
CONTROLLED OUTPUT PREVIEW
Obelisk material visualization interface showing an Inconel 718 conceptual structure

Real interface capture. Conceptual structure shown; not presented as a solver-relaxed commercial alloy unit cell.

04 / INITIAL FOCUS

One focus.
High-temperature aerospace.

Obelisk is narrowing its first operating domain to materials where temperature, time, atmosphere, processing, and microstructure interact under severe constraints.

01High-temperature superalloys
02Creep and fatigue resistance
03Oxidation and hot corrosion
04Thermal-barrier systems
05Heat-treatment strategy
06Manufacturing-route evaluation

05 / CASE STUDY

KNOWN-MATERIAL EVALUATION

Inconel 718

An example known-material analysis connecting chemistry, processing, failure mechanisms, and a proposed test plan. This is a qualitative review example, not an independent accuracy benchmark.

ANALYSIS / QUALITATIVEProcessing and failure-mechanism summary for technical review
VISUAL / CONCEPTUALIllustrative atomic model, not a validated alloy microstructure
IDENTIFIEDNi-based precipitation-strengthened superalloy
PROCESSING LOGICVacuum melting, thermomechanical processing, solution treatment, two-step aging
KEY MECHANISMSγ″ / γ′ strengthening, phase coarsening, creep, oxidation, fatigue
VALIDATION PATHMechanical testing, microscopy, chemistry mapping, phase confirmation, exposure testing

Known-material agreement demonstrates workflow behavior. It does not establish universal predictive accuracy or replace qualified engineering review and physical testing.

06 / SCIENTIFIC INTEGRITY

Prediction is not proof.
Obelisk makes the distinction visible.

Each technical claim should retain its origin, confidence, limitations, and next validation action. The platform is built to expose uncertainty, not decorate it.

Evidence labels and access boundaries +
01Referenced
02Calculated
03Inferred
04Predicted
05Unverified
06Requires testing
PUBLIC SURFACE

Technical context without engine access.

Selected case studies, methodology, limitations, and demonstration requests.

CONTROLLED ACCESS

Qualified demonstrations and pilot evaluations.

The scientific engine, customer workspaces, candidate data, and internal logic remain private.

DATA BOUNDARY

No confidential submissions through this form.

Protected project information requires a separate agreement and controlled workspace.

FREE GUIDED EVALUATION / OPERATOR-RUN

Your questions.
A focused technical handoff.

We operate Obelisk for your team and deliver the results. No installation or hosted app access required.

DURATIONTwo weeks from the agreed start date
SCOPEUp to three agreed, non-confidential aerospace materials questions
DELIVERABLESSource-linked reports covering material options, manufacturing routes, degradation risks, and proposed validation tests
SUPPORTOne walkthrough call and one revision
PRICINGFree two-week evaluation. Continued engagements priced separately.
BEFORE WE STARTAgree on questions, deliverables, timing, and what success looks like

Research assistance, not certified engineering conclusions or guaranteed discoveries. Physical testing and heavy simulations are excluded unless separately agreed. No payment is required for the agreed evaluation. Continuing afterward is optional and requires a separate paid agreement.

TECHNICAL Q&A / MATERIALS INTELLIGENCE

The intelligence
behind the workflow.

What Obelisk does, how it helps, and what your team receives.

Discuss your materials question
DIRECT CONTACT

How do we contact you?

Oumar.obelisk@gmail.com

Email us for a free evaluation, technical questions, or collaboration. Please keep your first message non-confidential.

ENGINEERING / EVALUATION / ACCESS13 QUESTIONS
01How does Obelisk reason about how a material could be manufactured?

Obelisk examines possible ways to make a material and explains why a route might fit. For an alloy, it can compare melting, forging, and heat treatment, then point out processing risks.

02How does it connect elemental chemistry to material behavior?

Obelisk looks at what each element contributes and how the elements work together. For example, it can explain why a change might improve heat resistance but make an alloy harder to manufacture.

03How does it generate proposed compositions and atomic structures?

You provide the properties and constraints you need, and Obelisk proposes compositions and structural models to investigate. Where coordinates are available, you can inspect the atomic arrangement; illustrative models are labeled separately from measured or calculated structures.

04How does it assess whether a proposed structure is physically plausible?

Obelisk checks for obvious problems, such as atoms placed too close together or a composition that does not fit the proposed material. These checks help reject bad inputs, but do not prove that a structure will be stable.

05How does it identify potential degradation and failure mechanisms?

You describe where and how the material will be used, and Obelisk examines what could go wrong. For example, a hot turbine environment may call for checking slow deformation under load, oxidation, and damage from repeated heating and cooling.

06How does it compare candidates with competing engineering requirements?

Obelisk compares candidates against your requirements, such as heat resistance, cost, and ease of manufacturing. It explains where each candidate performs better and what you would give up by choosing it.

07How does it redesign a candidate to address toxicity or processing constraints?

You can ask for a change, such as reducing toxicity or making a candidate easier to manufacture. Obelisk proposes alternatives and explains how those changes could affect performance.

08How does it distinguish an established manufacturing route from an experimental proposal?

Obelisk separates methods supported by references from routes it proposes for investigation. It also distinguishes how the material was made, so a property reported for a forged part is not automatically treated as valid for a 3D-printed one.

09How does it select relevant calculations and validation tests?

Obelisk suggests checks that address the question you are asking, such as strength testing, imaging, or oxidation testing. It distinguishes a recommended test from a completed calculation; not every answer uses DFT.

10How does it turn its analysis into an engineering dossier?

Obelisk organizes the analysis into a technical report you can review and share. It brings together material comparisons, manufacturing options, risks, available sources, assumptions, and suggested next tests.

11How does it carry material context into follow-up questions?

You can ask about the same material without starting the explanation over. For example, after discussing heat resistance, you can ask how changing the manufacturing route might affect it; during the evaluation, we confirm the candidate and conditions being discussed.

12What does the free evaluation include?

The free evaluation lasts two weeks and covers up to three agreed, non-confidential aerospace materials questions. We run Obelisk for you and deliver reports, one walkthrough, and one revision; it does not include direct app access, physical testing, or heavy simulations unless separately agreed.

13How much does Obelisk cost?

The two-week guided evaluation is free within the agreed scope. If you want to continue, we discuss your requirements and agree on pricing privately before starting paid work. There are no automatic charges or renewals.

07 / CONTROLLED ACCESS

Bring one materials bottleneck.

Request a free, two-week guided evaluation, or join the free waitlist for future availability.

A waitlist signup is an expression of interest, not a purchase or immediate engine access. Select your request type in the form.

01

Requests are reviewed manually.

02

We contact you to agree on scope and availability. No automatic paid renewal.

03

Do not submit proprietary material data.

FREE EVALUATION / WAITLISTNO PAYMENT REQUIRED

Questions? Oumar.obelisk@gmail.com. Submit only non-confidential information. No automatic app access.