Autocoder

Texas Instruments Interview Questions and Answers for India Embedded Roles

Last updated 2026-11-19

Texas Instruments is known for analog chips and embedded processors, so a software or firmware interview there often mixes embedded C with signal-chain basics: sampling, ADC resolution, serial interfaces, timers and low-power design. This page gives twelve practice questions with worked answers in that mix, followed by what the rounds are commonly reported to look like.

These are practice questions for the topics such roles are commonly reported to cover, not copies of a real interview. For coding problems in the same style, see Texas Instruments coding round questions.

Typical rolesEmbedded software, firmware, applications and test engineering
Core topicsEmbedded C, interfaces, ADC and sampling, timers, interrupts, fixed-point
Rounds (commonly reported)Written or online test, technical interviews, HR

Rounds as commonly reported

Prep guides usually describe a screening test with technical MCQs from electronics and C, then one or more technical interviews that go through your resume, then an HR round. The details differ by team and year, and the company does not publish a fixed format, so check your own invitation.

Candidates from ECE and EEE are often asked analog and digital basics, while CSE candidates usually see C, data structures and operating-system questions. Whatever your branch, be ready to explain any microcontroller project on your resume down to the register or peripheral level.

Topics to revise

Hiring status

This site has no sourced hiring statements for Texas Instruments, so it does not quote any drive date, opening count or package. Do not trust posts that do without a link to the company.

Check the company's own careers site for current openings and application rules. For general preparation, see how to crack the technical interview.

Texas Instruments Interview Questions and Answers for India Embedded Roles: questions with worked answers

  1. 1

    State the sampling theorem. What happens if you sample an 8 kHz tone at 10 kHz?

    easy

    To reconstruct a band-limited signal, the sampling rate must exceed twice its highest frequency. Sampling at 10 kHz, the Nyquist frequency is 5 kHz, so an 8 kHz tone folds back (aliases) to |10 - 8| = 2 kHz and is indistinguishable from a real 2 kHz tone after sampling. The fix is an anti-aliasing low-pass filter before the ADC, or a higher sample rate.

  2. 2

    A 12-bit ADC has a 3.3 V reference. What is one LSB, and what voltage does a reading of 2048 represent?

    easy

    There are 2^12 = 4096 codes, so one LSB is 3.3 / 4096, about 0.806 mV. A reading of 2048 corresponds to 2048 x 0.806 mV, which is about 1.65 V, half the reference. Mention that noise, offset and gain error reduce the usable resolution (the effective number of bits).

  3. 3

    Compare UART, SPI and I2C.

    medium

    UART is asynchronous, point to point, uses TX and RX and needs both sides to agree on baud rate and frame format. SPI is synchronous, uses a clock, MOSI, MISO and one chip-select per slave, is full duplex and fast, but needs more pins. I2C is synchronous with two wires, SDA and SCL, supports many devices by 7-bit address on one bus, and uses open-drain lines with pull-up resistors, so it is slower and its speed is limited by bus capacitance and pull-up value. Choose by pin count, speed and number of devices.

  4. 4

    Why does I2C need pull-up resistors, and how does the value affect the bus?

    medium

    I2C devices only pull a line low, they never drive it high (open-drain), so a resistor to the supply makes the line high when nobody pulls it down. A smaller resistor gives faster rising edges but uses more current and requires devices to sink more; a larger one saves power but slows the rise time and can break timing at higher speeds or with long wires. Choose a value that meets the rise-time limit for the mode and the total bus capacitance.

  5. 5

    What is the gain of an inverting op-amp with Rin = 10 kohm and Rf = 100 kohm?

    easy

    With negative feedback and an ideal op-amp, the inverting input is a virtual ground, so the input current is Vin / Rin and it all flows through Rf. Vout = -(Rf / Rin) x Vin = -10 x Vin, a gain of -10. A 0.2 V input gives -2 V, provided the supply rails allow it. Real op-amps are limited by gain-bandwidth product and output swing.

  6. 6

    A timer runs from a 1 MHz clock with its period register set to 999 (counts 0 to 999), and the compare value is 250. What are the PWM frequency and duty cycle?

    medium

    The counter has 1000 states per period, so the frequency is 1 MHz / 1000 = 1 kHz. The output is high for 250 of the 1000 counts, so the duty cycle is 25 percent. Exact register conventions (whether the period is N or N - 1 counts, and whether the output is high before or after the compare) depend on the part, so mention you would confirm in the datasheet.

  7. 7

    What is a watchdog timer and how should firmware use it?

    medium

    A watchdog is a hardware timer that resets the microcontroller if software does not clear it within a set time. Firmware should clear it from the main loop only when all critical tasks have shown progress, not from a timer interrupt, which would keep the dog fed while the main code is hung. After a reset, read the reset-cause register and log that the watchdog caused it.

  8. 8

    How would you debounce a mechanical switch?

    medium

    Contacts bounce for a few milliseconds when they open or close, giving several edges. In hardware use an RC filter with a Schmitt trigger. In software, sample the pin at a fixed rate, such as every 1 to 5 ms, and accept a new state only after it has been the same for N consecutive samples. Do not use a long delay inside an ISR; do the sampling from a timer tick.

  9. 9

    Explain a safe way to pass data from an ISR to the main loop.

    medium

    Use a single-producer, single-consumer ring buffer: the ISR writes at the head index and the main loop reads at the tail, and each index is changed by only one side. Mark the indices volatile, make the capacity a power of two so wrapping is a mask, and decide what happens on overflow, either dropping data or overwriting. On cores where index updates are not atomic, briefly disable interrupts around the read side.

    c
    #include <stdio.h>
    #include <stdint.h>
    #define SIZE 8u                      /* power of two */
    static volatile uint8_t buf[SIZE];
    static volatile uint8_t head, tail;
    int put(uint8_t v) {                 /* producer, e.g. ISR */
        uint8_t next = (head + 1u) & (SIZE - 1u);
        if (next == tail) return 0;      /* full */
        buf[head] = v; head = next; return 1;
    }
    int get(uint8_t *v) {                /* consumer, main loop */
        if (head == tail) return 0;      /* empty */
        *v = buf[tail]; tail = (tail + 1u) & (SIZE - 1u); return 1;
    }
    int main(void) {
        for (uint8_t i = 1; i <= 9; i++) put(i);   /* only 7 fit */
        uint8_t v; int n = 0;
        while (get(&v)) n++;
        printf("%d\n", n); /* 7 */
        return 0;
    }
  10. 10

    How do you multiply two Q15 fixed-point numbers on a core with no floating-point unit?

    hard

    A Q15 value is a 16-bit signed integer representing x / 32768. Multiply as 32-bit integers, then shift right by 15 to return to Q15: 0.5 is 16384, and 16384 x 16384 >> 15 = 8192, which is 0.25. The only overflow case is -1 x -1 = +1, which does not fit in Q15, so saturate to 32767. Add rounding by adding 1 << 14 before the shift if the application needs it.

    c
    #include <stdio.h>
    #include <stdint.h>
    int16_t q15_mul(int16_t a, int16_t b) {
        int32_t p = ((int32_t)a * b) >> 15;
        if (p > 32767) p = 32767;     /* -1 * -1 case */
        return (int16_t)p;
    }
    int main(void) {
        printf("%d %d\n", q15_mul(16384, 16384), q15_mul(-32768, -32768)); /* 8192 32767 */
        return 0;
    }
  11. 11

    How do you reduce power in a battery-powered microcontroller design?

    medium

    Run at the lowest clock and voltage that meets the deadline, switch off unused peripherals and clocks, and spend most of the time in a low-power sleep mode with a wake-up from a timer or an interrupt. Write firmware as an event loop that sleeps when idle, instead of polling. Reduce activity too: batch radio or sensor transfers, use DMA so the core can sleep during transfers, and turn off pull-ups or floating pins that leak current.

  12. 12

    HR: Why Texas Instruments, and describe an embedded or electronics project you built.

    easy

    Choose a project with a real result, such as a sensor logger, a motor controller or a small RTOS, and explain the part you owned: the interface you used, a problem you debugged with a multimeter, scope or logic analyser, and what you would change. Link that to the company by naming a type of product you find interesting, rather than restating that it is a large firm. Also read how to crack the HR interview.

Frequently asked questions

Do Texas Instruments interviews include analog electronics for software roles?

Commonly reported for firmware and applications roles: expect sampling, ADC and op-amp basics next to embedded C. Pure software roles lean more on C, data structures and operating systems.

How should a CSE student prepare for an embedded interview?

Learn the basics of one microcontroller family, write a few drivers for UART, SPI or I2C, and revise volatile, interrupts and bit operations. Be ready to talk about a project at register level.

Are these real TI questions?

No. They are practice questions in the style of the topics such roles are commonly reported to cover. The company does not publish its interview questions.

Where can I check current TI openings?

Use the company's own careers site. Do not rely on posts quoting drive dates or packages without a source.

Related guides

More in Company Interview Questions

See all in this section