Applied physics / engineering systems / IP creation

Make the hard things computable.

We solve difficult engineering problems. We help people understand how things behave, improve designs, and build better engineering tools.

How we work

We use physics, computation, and software to turn difficult questions into clearer engineering decisions. Depending on the problem, that can involve numerical modelling, simulation, custom computational tools, optimisation, automation, or coupled physical analysis.

01 Physics first
02 Systems minded
03 Research active
01 — The firm

We work on difficult systems.

Synergy Dynamics works across applied physics, computational engineering, materials, electrical systems, and engineering software. The physical system comes first; computation and automation follow where they improve understanding, design, or execution.

We work across applied physics, computational engineering, materials chemistry, electrical systems, and engineering software. The starting point is the physical system; computation and automation follow where they are useful.

01Computational & Aerothermal Engineering
02Materials & Chemical Systems
03Electrical & Electrothermal Systems
+Engineering Intelligence
01

Understand

Define the system, constraints, assumptions, and engineering objective.

02

Compute

Build models, simulations, numerical methods, and analyses.

03

Build

Translate understanding into systems, workflows, tools, or prototypes.

04

Productize

Turn validated capability into reusable infrastructure, IP, or products.

02 — Internal platform / active R&D

Flowlab.
Less fragmented.

Engineering work often breaks into too many disconnected steps. Flowlab is an attempt to connect those steps so a difficult problem can move more clearly from question to decision.

Explore the technical idea

Flowlab is a proprietary engineering workflow platform under development. It explores how problem definition, models, computation, analysis, interpretation, and design iteration can be represented as one structured computational workflow rather than a sequence of manual handoffs.

Inspect the seven-stage architecture
01
Engineering problemDefine the question, objective, and physical system.
−
02
Models & constraintsMake assumptions, geometry, material properties, and boundaries explicit.
+
03
Simulation / computationConnect numerical models and engineering tools to the problem.
+
04
AnalysisTransform solver output into interpretable engineering evidence.
+
05
AI-assisted reasoningInvestigate where reasoning support can intervene without replacing physical judgment.
+
06
Design iterationUse structured results to explore the next engineering question.
+
07
Engineering decisionKeep the final decision grounded in assumptions, evidence, and constraints.
+

This is an active area of investigation. The platform is presented here as development work and a research direction, not as a finished autonomous engineering product.

In development · active R&D
03 — Research

Questions before claims.

Selected work shows the kinds of physical systems, computational methods, and engineering questions currently being investigated.

Aerothermal physics / 2025

How does a material change heat flow through a phone?

Active researchThermal study

We compared chassis and cover materials to understand how heat moves through a phone-like structure. The work combines simulation with a custom numerical check.

Paper figure comparing temperature differences across chassis and cover material combinations

Source figure: Mali, “Smartphone thermal synergy” (2026), material-combination comparison.

Explore the technical work

Methods: 2D steady-state finite element analysis in ANSYS, compared across chassis and cover materials, with a custom one-dimensional explicit finite volume method in Python for cross-checking. The supplied paper includes temperature fields, material comparisons, and solver validation.

Interpretation: This is simulated work. It is presented as an engineering investigation, not as experimental validation or a commercial product result.

DemonstratedPaper figure

Material-combination comparison extracted from the supplied research document.

Evidence statusSimulation

Computational result; experimental validation is not claimed here.

Aerothermal physics / 2026

Can bio-inspired nozzle surfaces make flow more uniform?

Active researchCFD study

We compared a conventional converging-diverging nozzle with a modified geometry inspired by natural flow-control structures. The question is whether changing the wall near the throat changes pressure behaviour.

Pressure-field comparison from the supplied CFD study, showing baseline and bio-inspired nozzle cases

Source figure: Mali, “Computational Fluid Dynamics Investigation of Bio-Inspired Flow Control Features in a Converging-Diverging Nozzle” (2026), pressure-field comparison.

Explore the technical work

Methods: OpenFOAM v13, incompressible turbulent flow, identical 50 m/s inlet velocity, and comparison of velocity and pressure distributions. The study reports pressure variation changing from 3954.48 Pa to 1979.22 Pa, approximately 49.95%, alongside an 8.26% reduction in peak velocity.

Evidence: The figures shown here are extracted from the supplied presentation of the CFD study. The paper reports no mesh-independence study, experimental validation, drag-force calculation, pumping-power evaluation, or efficiency metric.

Simulated resultPressure field

Baseline and bio-inspired cases retain their original paper labels and colour scales.

Read with careResearch stage

Comparative CFD evidence, not experimental validation or a commercial performance claim.

Electrical systems / software study

Can a circuit be described clearly enough for a computer to solve?

DemonstratedPython / NumPy

A small Python circuit solver turns a component list into a solvable electrical model. It is a focused study in making circuit relationships explicit, inspectable, and computable.

Explore the technical work

Method: The script accepts a netlist of resistors and independent voltage sources, including component names, node pairs, and values. It builds the conductance matrix, augments it with voltage-source constraints using Modified Nodal Analysis, and solves the resulting linear system with NumPy.

Scope: The current solver is a compact educational and computational project. The source reviewed here documents resistor and voltage-source elements; no broader component support is claimed.

View the solver source
SourcePublic code

Python implementation by Sneha Sahjwani.

ApproachModified Nodal Analysis

Matrix assembly for circuit equations and voltage-source constraints.

01
Phone thermal solver

First-principles 1D transient heat-diffusion solver used for the thermal study, with explicit finite volume formulation and analytical validation.

View FVM repository
02
Converging-diverging nozzle cases

OpenFOAM case files for the normal Rao nozzle and bio-inspired feature nozzle comparison.

View CFD cases
04 — Where we work

Start with the problem.

We help teams understand difficult systems, test possible improvements, and turn useful technical work into better tools.

01

Understand how systems behave

Engineering & simulation: CFD, FEA, thermal analysis, heat transfer, aerothermal analysis, and multiphysics modelling.

02

Make repeated work clearer

Engineering automation: workflows, computational pipelines, tool integration, simulation automation, and AI-assisted workflows.

03

Answer an uncertain question

Technical research: feasibility studies, first-principles analysis, computational investigations, and prototype computational tools.

04

Develop something new

Advanced engineering and R&D: multidisciplinary problems, new computational approaches, technology feasibility, and proprietary tools.

05 — Contact

Start a technical conversation.

For engineering, research, computational development, or collaboration inquiries, contact Synergy Dynamics.