What engineers do

Engineers use science and math to solve problems for people. A bridge that carries traffic, a water filter, a phone app, a prosthetic hand: each one started with someone who had a problem and an engineer who asked, "How could we fix that?"

Engineers almost never get to design whatever they like. They work within limits: a budget, a deadline, the materials available, safety rules, the space something has to fit in. Solving a problem well inside those limits is the heart of the job.

The design process is a loop

Different textbooks draw it with slightly different boxes, but the engineering design process as it's commonly taught looks like this:

1. Define the problem. Who is it for? What must the design do? What limits do you have to work within?
2. Research. How have other people solved similar problems? What do you need to know or measure?
3. Brainstorm. Come up with many ideas, including odd ones. Don't judge yet.
4. Choose. Compare your best ideas against what the design needs to do, and pick one.
5. Build a prototype. Make a working version, often a rough one, so you can test it.
6. Test. Check the prototype against your requirements and write down what happens.
7. Improve. Fix what failed, then test again.
8. Communicate. Share the design, your results and your reasons so others can use or build on it.

It's a loop, not a straight line. Testing usually sends you back to improve, and sometimes all the way back to redefine the problem. Each pass through the loop is called an iteration.

Criteria, constraints and trade-offs

Defining the problem means writing down two kinds of things:

• Criteria are what a good design should do, the qualities you use to judge and compare ideas: how stable, how easy to use, how good it looks.
• Constraints are limits the design must stay within, no matter how good it is otherwise: cost, time, materials, size and safety.

An idea that scores well on every criterion but breaks a constraint is out. A $50 design doesn't count if your budget is $5.

You rarely get the best of everything. A sturdier design may cost more or take longer to build. Giving up some of one thing to get more of another is a trade-off, and engineers make them all the time.

A decision matrix helps you make trade-offs on purpose. List your criteria, give each a weight for how much it matters, and score each idea (say, 1 to 5) on every criterion. Multiply each score by its weight and add them up. The highest weighted total is a strong candidate, not an automatic winner, but it shows clearly why you chose what you did.

Failure is information

In engineering, a failed test isn't a bad grade. It's data. It tells you exactly where your design and the real world disagree, which is the most useful thing you can learn early, while changes are cheap.

That's why engineers build prototypes, often rough ones from cardboard, tape or scrap material. It's far better to find out a stand tips over when it's made of scrap wood than after you've made a hundred of them.

When something fails, ask three questions: What happened? Why did it happen? What one change would fix it? Change one thing at a time, then test again. If you change five things at once and it works, you won't know which change mattered.

Keep an engineering notebook

An engineering notebook is a dated record of your whole design: the problem statement, criteria and constraints, sketches of every idea (even rejected ones), your decision matrix, measurements, test results and what you changed after each test.

Write in it as you go, not afterward. Date each page, label your sketches, and never tear out pages. Cross mistakes out with a single line so they stay readable. A failed idea you recorded can save you, or a teammate, from trying it again.

The notebook also makes the last step, communicating, much easier. When someone asks "Why did you build it that way?", the answer is already written down.

The major fields

Every field of engineering uses the design process from Lesson 1. What changes is what they design. Here are the big ones, each with something you've probably used this week:

• Mechanical engineers design things that move or use energy and heat: engines, pumps, robots, refrigerators. Example: the gears and brakes on a bicycle.
• Electrical engineers work with electricity, from power lines to tiny circuits. Example: the charger that safely turns power from the wall into what your phone can use.
• Civil engineers design roads, bridges, tunnels, dams and water systems. Structural engineering is a specialty within civil that makes sure buildings and bridges carry their loads safely. Example: the road you take to school.
• Chemical engineers turn raw materials into useful products by designing large-scale chemical processes. Example: the factory process that makes toothpaste by the ton.
• Computer engineers design computer hardware and the place where hardware meets software, like chips and circuit boards. Software engineers design, build and test programs. Example: the apps on your phone and the chip that runs them.
• Biomedical engineers combine engineering with biology and medicine. Example: a hearing aid, or a prosthetic (artificial) leg.
• Aerospace engineers design aircraft, spacecraft and satellites. Example: the wings of a passenger airplane.
• Environmental engineers protect people and nature by dealing with water, air, soil and waste. Example: the treatment plant that makes your tap water safe to drink.
• Industrial engineers make systems and processes work better, faster and more safely. Example: how a warehouse picks, packs and ships orders without wasted steps.

Big projects need many fields

Real projects rarely belong to one field. Think about a hospital. Civil and structural engineers design the building and its foundation. Mechanical engineers handle heating, cooling and fresh air. Electrical engineers plan the power, including backup power for when the grid goes down. Biomedical engineers work on the medical equipment, software engineers build the patient-record systems, and industrial engineers plan how patients and staff move through it all.

A phone needs electrical and computer engineers for the circuits, software engineers for the apps, mechanical engineers for the case and for keeping it cool, chemical engineers for battery materials, and industrial engineers to build millions of them. A bridge needs structural engineers, plus environmental engineers protecting the river underneath.

That's why teamwork matters: you'll often work with people who know things you don't.

Engineers, technicians and technologists

Engineering teams include more than engineers. In general:

• Engineers focus on designing and analyzing: deciding what to build and using math and science to predict whether it will work.
• Engineering technologists usually study engineering technology, which is more hands-on and applied. They often put designs into practice, improve processes and run tests.
• Engineering technicians often train through a shorter program, such as a two-year associate degree. They build, test, measure, repair and maintain equipment, and collect the data engineers rely on.

Titles and duties vary by employer, and every role matters: a design is only as good as the people who build, test and maintain it.

Degrees and licensure

Most engineers earn a four-year bachelor's degree in engineering. Whatever the field, you can expect math through calculus and beyond, physics, chemistry, design projects, programming and lab courses where you test real things. Many programs share a common set of first-year courses before you specialize.

In the US, look for programs accredited by ABET, an organization that reviews engineering programs to check that they meet quality standards. It matters later if you want a license.

A Professional Engineer (PE) license lets an engineer take legal responsibility for work that affects the public, such as signing off on plans. Each state sets its own rules, but the path generally looks like this:

1. Earn a degree from an accredited engineering program.
2. Pass the Fundamentals of Engineering (FE) exam, often near or just after graduation.
3. Gain several years of work experience under licensed engineers.
4. Pass the Principles and Practice of Engineering (PE) exam in your field.

Licensure is especially common in civil engineering and related fields, where engineers approve public projects like roads and water systems. Many engineers in other fields, especially those designing products inside companies, work without one.

Skills for every field, and how to start now

Some skills matter no matter which field you pick: problem solving, math, communication, teamwork and writing. Engineers write reports, emails, test results and instructions all the time, and a great design that no one understands never gets built.

You don't have to wait for college to explore. In high school you can:

• Take the most challenging math and science classes you can handle, plus any design, programming or shop classes offered.
• Join a robotics or engineering club, or start one with a teacher's help.
• Enter a science fair with an engineering project.
• Look for summer engineering programs; many offer financial aid.
• Ask about job shadowing: spending a day following an engineer at work.
• Do an informational interview: a short conversation where you ask someone about their job.
• Build things at home, safely. Wear safety glasses, have an adult supervise any tools, and stick to small batteries for anything electrical. Never experiment with wall outlets or household wiring.

Why units matter

If someone tells you a board is "3 long", you can't cut it. Three what? Three millimeters, three feet and three meters are completely different boards. A number without a unit is meaningless. Always write the unit, even in the middle of a calculation.

Mixing units is one of the most common ways real engineering goes wrong. When one team works in inches and pounds and another assumes centimeters and kilograms, every number can look reasonable while the parts don't fit, a load is calculated wrong, or a system behaves in a way nobody planned. Mistakes like this have ruined expensive projects. The fix is simple: agree on units up front, label everything, and convert carefully when you have to.

SI units and prefixes

Scientists and engineers around the world use the International System of Units (SI), the modern metric system. It has seven base units. These five are the ones you'll use most:

• Meter (m) for length.
• Kilogram (kg) for mass.
• Second (s) for time.
• Ampere (A) for electric current, which you'll meet in Lesson 5.
• Kelvin (K) for temperature. A change of 1 K is the same size as a change of 1 °C, and 0 °C is exactly 273.15 K, so room temperature of about 20 °C is about 293 K.

(The other two are the mole, for amount of substance, and the candela, for brightness of light.)

Prefixes scale a unit up or down by powers of ten, and these conversions are exact:

• milli- (m) means one thousandth: 1 mm = 0.001 m, so 1 m = 1,000 mm.
• centi- (c) means one hundredth: 1 cm = 0.01 m, so 1 m = 100 cm.
• kilo- (k) means one thousand: 1 km = 1,000 m, and 1 kg = 1,000 g.
• mega- (M) means one million: 1 megawatt (MW) = 1,000,000 watts.

In the US you'll also see US customary units such as inches, feet, pounds and miles. Engineers here often use both, so conversion is a core skill.

Converting units

A conversion factor is a fraction whose top and bottom are equal amounts in different units, like 2.54 cm over 1 in. Because the top equals the bottom, multiplying by it changes the units without changing the actual size. Some useful ones:

• 1 inch = 2.54 cm (exact, by definition)
• 1 kg ≈ 2.205 lb
• 1 mile ≈ 1.609 km

To convert, write the value with its unit, then multiply by the factor arranged so the unit you don't want cancels:

1. Start with what you have: 150 lb.
2. Pick the factor with pounds on the bottom so they cancel: 1 kg / 2.205 lb.
3. Multiply: 150 lb × (1 kg / 2.205 lb) = 150 ÷ 2.205 kg ≈ 68.03 kg, or about 68 kg.

The same method turns a 12-inch ruler into centimeters: 12 in × (2.54 cm / 1 in) = 30.48 cm. If the units don't cancel the way you want, the factor is upside down.

Precision, accuracy and percent error

These two words sound alike but mean different things. Accuracy is how close a measurement is to the true value. Precision is how close repeated measurements are to each other. Picture darts: a tight cluster far from the bullseye is precise but not accurate. Darts scattered evenly around the bullseye are accurate on average but not precise. Good measurements are both.

Every measurement has some error, the difference between what you measured and the true, or accepted, value. Error isn't a blunder; it comes from the tool's limits, reading a scale at an angle, or the object itself. Because no part can be made perfectly either, drawings give a tolerance: a part marked "10.0 mm ± 0.1 mm" is acceptable anywhere from 9.9 mm to 10.1 mm. A part measuring 10.08 mm passes; one measuring 9.85 mm is rejected.

To compare errors fairly, engineers use percent error:

percent error = |measured − accepted| ÷ accepted × 100

The bars mean "absolute value", so percent error is never negative. Being off by 1 cm on a 10 cm pencil is a 10% error; being off by 1 cm on a 100 m track is only 0.01%. Try the calculator below with your own measurements.

Estimation and sanity checks

Engineers often need a rough answer fast, before any careful calculation. A back-of-the-envelope or Fermi estimate breaks a big question into pieces you can guess sensibly, then multiplies them out using round numbers. The goal is the right size, not the exact value.

How many times does a car tire turn in one mile? A typical car tire is about 0.65 m across, so its circumference is about π × 0.65 m ≈ 2 m. A mile is about 1,600 m. So 1,600 m ÷ 2 m per turn ≈ 800 turns. A careful calculation gives about 790, so the estimate is close.

Estimates also help you sanity-check answers. Before you trust a result, ask:

• Does it make sense? A desk 1.2 km long or a backpack weighing 3 g means something went wrong.
• Do the units work out? If you're finding a length and your answer comes out in kilograms, look for the mistake.
• Is it close to your estimate? If your careful answer is 100 times your rough one, check for a misplaced prefix like milli- or kilo-.

Force and the newton

A force is a push or a pull. In SI units it's measured in newtons (N). One newton is the force that speeds up a 1 kg mass by 1 meter per second, every second. A 100 g apple in your hand pushes down with about 1 N.

Weight is the pull of gravity on a mass. On Earth, that's about 9.8 N for every kilogram:

weight (N) ≈ mass (kg) × 9.8 m/s²

A 1 kg bag of rice weighs about 9.8 N. A 60 kg person weighs about 588 N. In US customary units, force is measured in pounds-force (lbf), and 1 lbf is about 4.45 N.

Loads, tension and compression

The forces a structure carries are called loads. There are two broad kinds:

• Dead load is the weight of the structure itself and anything permanently attached, like beams, floors and walls. It's always there.
• Live load is anything that comes and goes: people, furniture, cars on a bridge, snow on a roof. Designers plan for the heavy cases.

Inside a structure, loads become two basic kinds of force. Tension pulls a part apart, like a rope in a tug-of-war. Compression squeezes it, like a column holding up a porch roof. Stress is force spread over an area, so the same force on a thinner part means more stress.

Triangles, trusses and beams

Pin four sticks into a square and it folds into a diamond with almost no push. Pin three into a triangle and it won't budge: with three side lengths fixed, a triangle can only have one shape, so it can't deform unless a side stretches or squashes. That's why triangles are called rigid.

A truss is a frame made of connected triangles. You'll see them in bridges, roof frames and cranes. Each member is mostly in tension or compression, so a truss carries big loads with little material.

A beam carries loads across a gap, like a shelf or a plank over a stream. Picture a beam resting on a support at each end with a load on the middle. As it sags, the top gets squeezed (compression) while the bottom gets stretched (tension). Bend a rubber eraser and you can see it happen.

Materials and the factor of safety

Choosing a material means trading off several properties:

• Strength: how much stress it can take before it breaks or permanently bends.
• Stiffness: how much it resists bending or stretching. Glass is stiff, yet it shatters easily.
• Weight: lighter matters for anything that moves or has to be carried.
• Cost, and how easy it is to cut and join.

Materials can behave differently in tension and compression. Steel is strong in tension. Concrete is strong in compression but weak in tension: a plain concrete beam cracks along its stretched bottom. So builders cast steel bars into the concrete where tension will be. That's reinforced concrete: each material does what it's good at.

Engineers never design a part to hold only the load they expect. They build in a factor of safety: the load that would make it fail divided by the expected load. A shelf expected to hold 200 N that would only fail at 600 N has a factor of safety of 3. The margin covers surprises: heavier loads, flaws in the material, wear over time. For real structures, building codes set how big it must be.

Levers and mechanical advantage

A lever is a stiff bar that turns around a pivot point called the fulcrum. Seesaws and pry bars are levers. The force you apply is the effort; what you're moving is the load.

A force's turning effect is called its moment (or torque):

moment = force × distance from the fulcrum

In newtons and meters, moment is in newton-meters (N·m). A lever balances when the moments on each side are equal:

load × load distance = effort × effort distance

So the effort needed is load × load distance ÷ effort distance. The mechanical advantage, how many times the lever multiplies your force, is effort distance ÷ load distance for an ideal lever with no friction. The trade-off: your end moves farther than the load does, by the same ratio. You get an easier push, over a longer distance.

Try the calculator below. Any units work if both distances use the same one.

Safety first

Everything in this lesson uses small batteries (about 1.5 V to 9 V) and low-power parts, like those in a classroom electronics set. At that level, the circuits here are safe to build with care.

• Never experiment with wall outlets, household wiring, extension cords or anything that plugs into the wall. Household electricity can kill.
• Never experiment with large batteries, such as car batteries or the packs in power tools, e-bikes or scooters. They can deliver enough current to cause burns and fires.
• Work with an adult or teacher supervising, and wear safety glasses when you cut or strip wire.
• If a part, wire or battery gets hot, smells or swells, disconnect it and stop.

Charge, current, voltage and resistance

Atoms contain tiny particles with electric charge. Electrons carry negative charge, and in metals some are free to move. Charge is measured in coulombs (C).

• Current is how much charge flows past a point each second, measured in amperes (A), or amps. 1 A means 1 coulomb per second. Small circuits often use milliamps (mA): 1 A = 1,000 mA.
• Voltage is the push that drives charge around a circuit, measured in volts (V). A battery provides a voltage between its two terminals.
• Resistance is how strongly something opposes current, measured in ohms (Ω).

A handy picture is water in pipes. The battery is like a pump, voltage is like the water pressure it creates, current is like how much water flows each second, and resistance is like a narrow section of pipe that slows the flow.

The analogy has limits. Cut a water pipe and water pours out; cut a wire and the current just stops, because no charge spills out. And the battery doesn't fill empty wires: the charges are already in the wire, and it pushes them along. These are SI units, used everywhere; there's no US customary version of a volt or an amp.

Closed, open and short circuits

Current only flows around a complete loop. A closed circuit is an unbroken path from one battery terminal, through a device such as a bulb, and back to the other terminal. Break the loop anywhere and you have an open circuit: no current, nothing happens. A switch is just a neat way of opening and closing a circuit.

What the path is made of matters. Conductors, like copper, aluminum and other metals, let charge flow easily. Insulators, like plastic, rubber, glass and dry wood, don't. That's why wires are metal inside and wrapped in plastic.

A short circuit happens when a path with almost no resistance connects the battery's terminals directly, skipping the device. With almost nothing to limit it, the current gets very large. Wires and the battery heat up fast; they can burn you, melt insulation, and some batteries can leak, swell or catch fire. Never connect a bare wire straight across a battery, and never carry loose batteries in a pocket with coins or keys.

Ohm's law and power

For many parts, especially resistors, voltage, current and resistance are linked by Ohm's law:

V = I × R (voltage in volts = current in amps × resistance in ohms)

Rearrange it to find whichever one you're missing:

• I = V ÷ R: a 1.5 V cell across 30 Ω gives 1.5 ÷ 30 = 0.05 A, or 50 mA.
• R = V ÷ I: 9 V pushing 0.03 A means 9 ÷ 0.03 = 300 Ω.
• V = I × R: 0.01 A through 470 Ω needs 0.01 × 470 = 4.7 V.

Use amps, not milliamps, in the formula, and convert at the end. Power is how fast energy is used, measured in watts (W): P = V × I. Power in a resistor turns into heat, so every resistor has a power rating it shouldn't go over. Small resistors like the ones in classroom sets are often rated for 0.25 W.

Try the calculator below: enter a voltage and a resistance to get the current and the power.

Series and parallel circuits

In a series circuit, parts are connected one after another in a single loop. The same current flows through each part, and resistances add: R = R1 + R2. If one part breaks, the loop opens and everything stops.

In a parallel circuit, each part gets its own branch between the same two points. Every branch has the same voltage across it, and the branch currents add up to the total. Because charge has more than one path, the total (equivalent) resistance is lower than any single branch. For two resistors in parallel:

R = (R1 × R2) ÷ (R1 + R2)

Two equal resistors in parallel give exactly half of one. The lights in a house are wired in parallel, which is why switching off one lamp doesn't turn off the rest.

Why nearly every engineer writes code

You don't have to become a software engineer to use code. Today almost every engineering field relies on it in three big ways:

• Controlling machines. Thermostats, traffic lights, washing machines, factory robots and car brakes all run small programs that decide what to do next.
• Analyzing data. A civil engineer might study readings from sensors on a bridge; a biomedical engineer might look at heart-rate data from a medical device. Code sorts through thousands of numbers far faster than a person can.
• Simulating designs. Engineers test a bridge, an airplane wing or a circuit on a computer before building it, so they can try many ideas cheaply and safely.

Whatever field you choose from Lesson 2, a little coding will make you a stronger engineer.

Algorithms and pseudocode

An algorithm is a set of step-by-step instructions for solving a problem. A recipe is an algorithm. A computer follows its instructions exactly, with no common sense to fill in gaps, so every step has to be clear.

Almost every algorithm is built from three structures:

• Sequence: do steps in order, one after another.
• Decisions: if something is true, do one thing; otherwise, do something else.
• Loops: repeat steps, either a set number of times or until something changes.

Before writing real code, engineers often write pseudocode: plain-language steps that look a bit like code but don't follow any programming language's rules. Here's pseudocode for a fan that cools a small greenhouse:

1. Repeat forever:
   1. Read the temperature.
   2. If temperature ≥ 28 °C, turn the fan on.
   3. Otherwise, turn the fan off.
   4. Wait 1 minute.

It uses all three structures. Once the logic is right, turning it into real code, in a free block-based coding tool or a text language, is much easier.

Sensors, microcontrollers and data

Many engineering projects follow the same pattern: input → processing → output.

• Input: a sensor measures something in the world, like temperature, light, distance or moisture, and turns it into an electrical signal.
• Processing: a microcontroller, a tiny computer on a small board, runs your program and decides what to do with that reading.
• Output: something happens, such as a light turning on, a motor spinning or a number appearing on a screen.

Sensors also collect data. Single readings are noisy, so engineers usually take several and find the average (mean): add them up and divide by how many there are. Greenhouse readings of 21, 23, 22, 24 and 20 °C add up to 110, and 110 ÷ 5 = 22 °C.

Watch for readings that don't fit. If the sensor sat in direct sun for a moment and read 55 °C, the average of all six readings jumps to 165 ÷ 6 = 27.5 °C, even though the room never got that warm. Look for patterns, too: a graph with time along the bottom and the reading up the side makes trends easy to spot, like soil slowly drying over a week or a sudden spike that points to a faulty reading.

Testing: expected vs. actual

Testing code is the "test" step of the design process. You write test cases: specific inputs, each with the expected result you work out before running anything. Then you run the program and record the actual result. If they match, the test passes. If not, you've found a bug, and now you can fix it.

Don't just test typical values. Bugs love to hide at the boundaries, so include edge cases: values exactly at a limit, just above and just below it, and the smallest and largest values the sensor can give.

Debugging, iteration and good records

Debugging is finding and fixing bugs, and it works best as a calm, step-by-step process:

1. Reproduce the problem, so you can make it happen on purpose.
2. Isolate it: narrow down which part of the program or which input causes it.
3. Fix one thing at a time, so you know which change made the difference.
4. Retest, including the tests that already passed, to be sure the fix didn't break something else.

This is iteration, the same loop as the design process from Lesson 1: define the problem, plan, build, test, improve, and repeat. Code is rarely right the first time, even for professionals.

Version history means saving numbered versions with a short note on what changed, so you can go back if a change makes things worse; many coding tools do this for you. Documentation means comments in your code and notes that explain what it does and why, so a teammate, or you in three months, can understand it.