Hit Refresh : A story of purposeful resets | Yogesh Kulkarni | TEDxCCOEW
Change is an accelerating force, particularly due to AI and technology, requiring professionals to embrace frequent, non-linear "resets" rather than expecting a smooth career trajectory. The speaker uses his own career path—moving from mechanical engineering to CAD to starting a company, pursuing a PhD, mastering machine learning, and becoming an AI advisor—to illustrate the necessity of these dramatic, non-incremental shifts. He advises that individuals must combine their core domain expertise with AI as a tool, aiming to become a specialized "rider on the wave."
## Speakers & Context
- Speaker: Unnamed professional, mechanical engineer by training.
- Context: Addressing a new generation about professional careers amidst accelerating technological change, specifically AI.
- Observation: Previous generations could expect a single job from start to retirement.
- Career Pattern: The speaker’s career has been characterized by multiple, deliberate "resets" or plunges.
## Theses & Positions
- Change is not just constant, but it is *accelerating*, especially concerning technology and artificial intelligence.
- Traditional career paths (stable, incremental progression) are obsolete due to external factors like AI.
- Professional success requires embracing multiple "resets"—dramatic, non-linear changes—rather than gradual improvements.
- AI should be treated as a *tool* to solve problems, not as the primary domain itself; one must anchor oneself in a core domain (e.g., mechanical, legal, artistic).
- Advancement in the career should follow a "sawtooth wave," always trending up and never letting the momentum down.
## Concepts & Definitions
- **Machine Design:** The process of designing physical objects, such as automobiles or chairs, by determining dimensions, material, and thickness.
- **CAD (Computer-Aided Design):** Performing designs using software instead of physical prototyping (e.g., using software to design a chair before manufacturing).
- **CAE (Computer-Aided Engineering):** Testing designs created in CAD using specialized software.
- **Defeature:** The process of simplifying a shape in a design (e.g., removing small holes, fillets, or chamfers) for subsequent structural testing.
- **Supervised Machine Learning:** An approach where logic or patterns (like "If the radius is less than five millimeters, remove it") are derived from hundreds of existing examples rather than explicitly coded by hand.
- **AI (Artificial Intelligence):** The underlying field being adopted as a mandatory skill across all professions.
- **Gartner Hype Cycle:** A model describing how technology adoption proceeds through hype, peak, decline, and eventual maturity.
- **Orbital Change:** The necessary dramatic shift in career focus, contrasting with small, incremental changes (like 10th or 5th percentile gains).
## Mechanisms & Processes
- **Mechanical Engineering Design Example:** Moving from physical prototyping to digital simulation using CAD and CAE.
- **PhD Pursuit Challenge:** The speaker found it impossible to maintain full group responsibilities while simultaneously conducting meaningful PhD research, necessitating a career break.
- **AI Discovery:** The speaker realized machine learning bypassed the need for hard-coded rules in "Defeaturing" by analyzing patterns across hundreds of examples.
- **AI Skill Application:** Recommends three functional roles for using AI:
1. **User:** Simple interaction via prompt engineering.
2. **Developer:** Building applications using AI.
3. **Researcher:** Innovating by developing new things, ideally requiring strong mathematics.
- **AI Integration Strategy:** The process is to take an AI-related technology, test if it solves a problem within the core domain, and if not useful, discard it; the core domain expertise remains.
## Timeline & Sequence
- **Bachelors:** Studied mechanical engineering in Pune.
- **Masters:** Studied CAD in the US.
- **First Reset:** Left a secure job to join a US-based startup office in China.
- **Career Progression:** Advanced from software engineer to lead, manager, and group manager within the CAD industry.
- **Second Reset:** Left stable employment to pursue a PhD at the College of Engineering Pune.
- **Third Reset:** Left PhD research/employment to focus on Machine Learning when the academic path proved too conflicting with professional responsibility.
- **Fourth Reset:** Left a stable job to become a solo AI advisor, working across domains like finance.
## Named Entities
- **Pune** — Location where the speaker completed his Bachelor's degree.
- **College of Engineering Pune** — Institution where the speaker enrolled for his PhD.
- **China** — Location of the startup where the speaker was the first employee.
- **US** — Origin of the multinational parent company and initial master's education.
## Numbers & Data
- Experience level: **20 plus years** in CAD domain.
- Age when starting machine learning career: **40+**.
- Age when becoming an AI advisor: Implied later than 40+.
- Core design stress test: **100kg, 120kg** occupants.
## Examples & Cases
- **Chair Design:** Basic function is dimensioning, material selection, and thickness calculation to withstand specified loads (100kg, 120kg).
- **Startup Start:** Beginning with "No HR, No IT," using only two tables in someone else's office.
- **PhD vs. Job:** The conflict between maintaining professional group responsibilities and executing meaningful research.
- **Defeature Example:** The need to simplify a complex shape for testing; rule example: *"If the radius is less than five millimeters, remove it."*
- **AI Insight Example:** A paper that solved "Defeature" by presenting hundreds of examples, leading to the realization of ML.
- **Career Growth:** The journey from "cushy job" to "first employee" in a startup, then to a "CAD giant" role, and finally to "solo" advisor status.
## Tools, Tech & Products
- **CAD (Computer-Aided Design)**
- **CAE (Computer-Aided Engineering)**
- **Software:** Used for design and testing simulations.
- **AI** (Artificial Intelligence)
- **Machine Learning:** The mathematical process providing pattern recognition beyond explicit coding.
- **Prompt Engineering:** A specific skill used when interacting with AI as a user.
## References Cited
- *Hit Refresh* — A famous book title used as a source for the talk's concept.
- *Tarang* — The Sanskrit word for "Wave," cited as the theme of the TEDx event.
## Counterarguments & Caveats
- The initial difficulty in articulating the value of storytelling (though this topic is not the main focus, the comparison is noted in the speaker's general context).
- The risk of getting stuck in a domain because "once you are in a company you are sort of bound by what that company does."
- The danger of assuming change must be incremental (e.g., "Plus 10th percentile, 5th percentile. That’s not the change.").
## Methodology
- **Career Retrospection:** Analyzing the speaker's own life as a case study to explain professional change.
- **Conceptual Modeling:** Using the Gartner Hype Cycle and the "sawtooth wave" to explain the dynamics of technological adoption and personal career management.
## Conclusions & Recommendations
- The theme suggests being a rider on the wave (*Tarang Savar*) or a friend to the wave (*Tarang Mitr*) rather than opposing it.
- The primary advice is to always be ready to "Hit Refresh" on one's career, suggesting that the necessary changes must be dramatic, "orbital change."
- Professionals must integrate AI as a supportive tool into their existing expertise rather than adopting AI as a sole focus.
## Implications & Consequences
- The consequence of ignoring AI is falling behind the acceleration curve, leading to stagnation in professional value.
- The positive consequence of adopting the "combo" (Domain Expertise + AI) is becoming specialized in a way that knowledge cannot easily replicate.
## Verbatim Moments
- *"change is not just constant, but it’s accelerating"*
- *"over the career of 20 years, you would have changed 5-6 jobs, right?"*
- *"you do it on computers, you do it in software."*
- *"So that was my first reset. First reset in the career."*
- *"I decided to leave my cushy job, and started to join, or decided to join the startup."*
- *"I should take my second plunge. I decided to leave the job and become a student again."*
- *"This is what is known as machine learning, supervised machine learning."*
- *"At 40+ age, nobody will give me a job with that seniority with that salary package. Not affordable."*
- *"it is becoming mandatory, compulsory to know at least something about AI."*
- *"AI is actually a tool. You have to have your own domain."*
- *"You have to have your own domain. If you are a mechanical engineer, be a mechanical engineer."*
- *"be a rider on the wave - Tarang Savar, or Tarang Mitr - be a friend of the wave, not oppose it because you don’t have a choice."*
- *"You have to have dramatic change, orbital change."*
- *"I would summarize my talk with another borrowed line from another tag line... Hit “Refresh” to refresh your career, at least for the newer generation."*