Tesla recently posted nearly 30 jobs and 2 internships related to Dojo. Most of the Tesla Dojo positions are in Palo Alto, California. Tesla posted one Dojo-related job in Texas and another in Colorado.
Tesla is looking for a Sr. DFT Verification Engineer and Sr. DFT Engineer in Austin, Texas. The Dojo team is looking for a Staff Physical Design Engineer in Fort Collins, Colorado.
Besides the two jobs in Texas, Tesla’s Dojo team is also searching for a few people to fill senior positions in Palo Alto, California, including a Sr. Site Reliability Engineer, Sr. Design Verification Engineer, and Sr. Firmware Engineer.
Tesla also wants to welcome interns to the Dojo team for the summer of 2025. The company is specifically looking for Performance Modeling Engineers and future Technical Program Managers.
Performance Modeling Engineer Internship Description
This position is expected to start around May 2025 and continue through the Summer term (approximately August 2025) or into Fall 2025 if available and there is an opportunity to do so. We ask for a minimum of 12 weeks, full-time and on-site, for most internships. Our internship program is for students who are actively enrolled in an academic program. Recent graduates seeking employment after graduation and not returning to school should apply for full-time positions, not internships.
International Students: If your work authorization is through CPT, please consult your school on your ability to work 40 hours per week before applying. You must be able to work 40 hours per week on-site. Many students will be limited to part-time during the academic year.
Location: Palo Alto, CA
As an intern on the Dojo Performance Modeling team, you will play an integral part in efficiently running Tesla’s neural networks on our in-house custom-silicon supercomputer system. You will be involved in tasks like running ML benchmarks to analyze and debug performance bottlenecks, develop new tests and build the infrastructure to automate these processes. We are looking for a motivated engineering student that is excited by the work Tesla is doing in pushing the envelope of real-world AI. The ideal candidate will have a strong background in computer architecture, analytical and cycle-based simulation, and AI workloads, with a passion for high-performance computing and complex systems modeling.
Performance Modeling Engineer Responsibilities
- Develop and validate microarchitecture simulations of a massively parallel machine for AI training, including system architecture, core architecture, memory hierarchy, and interconnects.
- Write, debug, and maintain robust infrastructure code for validating the Dojo performance.
- Create and maintain performance dashboards on the Dojo system.
- Collaborate with architects and engineers to understand the requirements of the simulation and ensure that it accurately models the behavior of the system.
- Develop and maintain software frameworks and tools to support testing and deployment.
- Participate in code reviews, testing, and debugging to ensure high-quality software.
Technical Program Manager (DOJO & AI Hardware) Internship Description
This position is expected to start around May 2025 and continue through the Summer term (approximately August 2025) or into Fall 2025 if available and there is an opportunity to do so. We ask for a minimum of 12 weeks, full-time and on-site, for most internships. Our internship program is for students who are actively enrolled in an academic program. Recent graduates seeking employment after graduation and not returning to school should apply for full-time positions, not internships.
International Students: If your work authorization is through CPT, please consult your school on your ability to work 40 hours per week before applying. You must be able to work 40 hours per week on-site. Many students will be limited to part-time during the academic year.
Location: Palo Alto, CA
Technical Program Manager (DOJO & AI Hardware) Internship Responsibilities
- Currently pursuing a degree in Mechanical, Electrical, Computer Science Engineering, or a related field
- Prior program management experience or managing a team, such as FSAE, Hyperloop, etc
- Desired to be proficient in Microsoft Office, JIRA, Confluence, and Git
- Experience in leading teams and proven ability to drive initiatives to conclusion
The Teslarati team would appreciate hearing from you. If you have any tips, contact me at maria@teslarati.com or via Twitter @Writer_01001101.
Tesla rolled out Full Self-Driving version 14.2 yesterday to members of the Early Access Program (EAP). Expectations were high, and Tesla surely delivered.
With the rollout of Tesla FSD v14.2, there were major benchmarks for improvement from the v14.1 suite, which spanned across seven improvements. Our final experience with v14.1 was with v14.1.7, and to be honest, things were good, but it felt like there were a handful of regressions from previous iterations.
While there were improvements in brake stabbing and hesitation, we did experience a few small interventions related to navigation and just overall performance. It was nothing major; there were no critical takeovers that required any major publicity, as they were more or less subjective things that I was not particularly comfortable with. Other drivers might have been more relaxed.
With v14.2 hitting our cars yesterday, there were a handful of things we truly noticed in terms of improvement, most notably the lack of brake stabbing and hesitation, a major complaint with v14.1.x.
However, in a 62-minute drive that was fully recorded, there were a lot of positives, and only one true complaint, which was something we haven’t had issues with in the past.
The Good
Lack of Brake Stabbing and Hesitation
Perhaps the most notable and publicized issue with v14.1.x was the presence of brake stabbing and hesitation. Arriving at intersections was particularly nerve-racking on the previous version simply because of this. At four-way stops, the car would not be assertive enough to take its turn, especially when other vehicles at the same intersection would inch forward or start to move.
This was a major problem.
However, there were no instances of this yesterday on our lengthy drive. It was much more assertive when arriving at these types of scenarios, but was also more patient when FSD knew it was not the car’s turn to proceed.
Can report on v14.2 today there were ZERO instances of break stabbing or hesitation at intersections today
It was a significant improvement from v14.1.x
— TESLARATI (@Teslarati) November 21, 2025
This improvement was the most noticeable throughout the drive, along with fixes in overall smoothness.
Speed Profiles Seem to Be More Reasonable
There were a handful of FSD v14 users who felt as if the loss of a Max Speed setting was a negative. However, these complaints will, in our opinion, begin to subside, especially as things have seemed to be refined quite nicely with v14.2.
Freeway driving is where this is especially noticeable. If it’s traveling too slow, just switch to a faster profile. If it’s too fast, switch to a slower profile. However, the speeds seem to be much more defined with each Speed Profile, which is something that I really find to be a huge advantage. Previously, you could tell the difference in speeds, but not in driving styles. At times, Standard felt a lot like Hurry. Now, you can clearly tell the difference between the two.
It seems as if Tesla made a goal that drivers should be able to tell which Speed Profile is active if it was not shown on the screen. With v14.1.x, this was not necessarily something that could be done. With v14.2, if someone tested me on which Speed Profile was being used, I’m fairly certain I could pick each one.
Better Overall Operation
I felt, at times, especially with v14.1.7, there were some jerky movements. Nothing that was super alarming, but there were times when things just felt a little more finicky than others.
v14.2 feels much smoother overall, with really great decision-making, lane changes that feel second nature, and a great speed of travel. It was a very comfortable ride.
The Bad
Parking
It feels as if there was a slight regression in parking quality, as both times v14.2 pulled into parking spots, I would have felt compelled to adjust manually if I were staying at my destinations. For the sake of testing, at my first destination, I arrived, allowed the car to park, and then left. At the tail-end of testing, I walked inside the store that FSD v14.2 drove me to, so I had to adjust the parking manually.
This was pretty disappointing. Apart from parking at Superchargers, which is always flawless, parking performance is something that needs some attention. The release notes for v14.2. state that parking spot selection and parking quality will improve with future versions.
Any issues with parking on your end? 14.1.7 didn’t have this trouble with parking pic.twitter.com/JPLRO2obUj
— TESLARATI (@Teslarati) November 21, 2025
However, this was truly my only complaint about v14.2.
You can check out our full 62-minute ride-along below:
Elon Musk
SpaceX issues statement on Starship V3 Booster 18 anomaly
The incident unfolded during gas-system pressure testing at the company’s Massey facility in Starbase, Texas.
SpaceX has issued an initial statement about Starship Booster 18’s anomaly early Friday. The incident unfolded during gas-system pressure testing at the company’s Massey facility in Starbase, Texas.
SpaceX’s initial comment
As per SpaceX in a post on its official account on social media platform X, Booster 18 was undergoing gas system pressure tests when the anomaly happened. Despite the nature of the incident, the company emphasized that no propellant was loaded, no engines were installed, and personnel were kept at a safe distance from the booster, resulting in zero injuries.
“Booster 18 suffered an anomaly during gas system pressure testing that we were conducting in advance of structural proof testing. No propellant was on the vehicle, and engines were not yet installed. The teams need time to investigate before we are confident of the cause. No one was injured as we maintain a safe distance for personnel during this type of testing. The site remains clear and we are working plans to safely reenter the site,” SpaceX wrote in its post on X.
Incident and aftermath
Livestream footage from LabPadre showed Booster 18’s lower half crumpling around the liquid oxygen tank area at approximately 4:04 a.m. CT. Subsequent images posted by on-site observers revealed extensive deformation across the booster’s lower structure. Needless to say, spaceflight observers have noted that Booster 18 would likely be a complete loss due to its anomaly.
Booster 18 had rolled out only a day earlier and was one of the first vehicles in the Starship V3 program. The V3 series incorporates structural reinforcements and reliability upgrades intended to prepare Starship for rapid-reuse testing and eventual tower-catch operations. Elon Musk has been optimistic about Starship V3, previously noting on X that the spacecraft might be able to complete initial missions to Mars.
Investor's Corner
Tesla analyst maintains $500 PT, says FSD drives better than humans now
The team also met with Tesla leaders for more than an hour to discuss autonomy, chip development, and upcoming deployment plans.
Tesla (NASDAQ:TSLA) received fresh support from Piper Sandler this week after analysts toured the Fremont Factory and tested the company’s latest Full Self-Driving software. The firm reaffirmed its $500 price target, stating that FSD V14 delivered a notably smooth robotaxi demonstration and may already perform at levels comparable to, if not better than, average human drivers.
The team also met with Tesla leaders for more than an hour to discuss autonomy, chip development, and upcoming deployment plans.
Analysts highlight autonomy progress
During more than 75 minutes of focused discussions, analysts reportedly focused on FSD v14’s updates. Piper Sandler’s team pointed to meaningful strides in perception, object handling, and overall ride smoothness during the robotaxi demo.
The visit also included discussions on updates to Tesla’s in-house chip initiatives, its Optimus program, and the growth of the company’s battery storage business. Analysts noted that Tesla continues refining cost structures and capital expenditure expectations, which are key elements in future margin recovery, as noted in a Yahoo Finance report.
Analyst Alexander Potter noted that “we think FSD is a truly impressive product that is (probably) already better at driving than the average American.” This conclusion was strengthened by what he described as a “flawless robotaxi ride to the hotel.”
Street targets diverge on TSLA
While Piper Sandler stands by its $500 target, it is not the highest estimate on the Street. Wedbush, for one, has a $600 per share price target for TSLA stock.
Other institutions have also weighed in on TSLA stock as of late. HSBC reiterated a Reduce rating with a $131 target, citing a gap between earnings fundamentals and the company’s market value. By contrast, TD Cowen maintained a Buy rating and a $509 target, pointing to strong autonomous driving demonstrations in Austin and the pace of software-driven improvements.
Stifel analysts also lifted their price target for Tesla to $508 per share over the company’s ongoing robotaxi and FSD programs.
Tesla Full Self-Driving v14.2 – Full Review, the Good and the Bad
SpaceX issues statement on Starship V3 Booster 18 anomaly
