Soft Skills Engineering - Episode 474: I hate the idea of firing a low performer and cheaper context switching

Episode Date: August 18, 2025

In this episode, Dave and Jamison answer these questions: Hi Dave & Jamison, Long time listener, first time google-form filler outer! I work in a hybrid role as a lead developer... and manager of a small team (less than 5). I’m new to management and most of ny experience so far has been with smart, motivated engineers. . . UNTIL! My new recruit is driving me crazy, they are clearly very capable, but just do not do the work. They are frequently late for work, frequently sign off early, and constantly evasive when I ask for updates. I have spoken to them about these issues a bunch, and everytime they are apologetic and say they “have some personal issues but are working on it” - and nothing changes. Urgh! I am pretty sure I will have to fire them, but I feel terrible about it! I know I can’t keep them on and pay them to do nothing, but what’s the best way to let somebody go? How do I break the news to the rest of the team? How do I avoid feeling bad for the rest of my life? Yours guiltily, Anon A listener named “erm what the sigma” asks, Do you have any advice on how to reduce the ramp-up time when context switching? I’ve always felt like context switching comes at a high cost for me—it just takes so long for me to mentally shift between tasks. This wasn’t much of a problem before, but I’ve recently become a tech lead and now my calendar is cluttered with meetings (why did I ask for this again??). I’m struggling to complete my coding stories because just as I hit my stride, I get pinged by someone on my team to help them or have to jump into yet another meeting. pls send help

Transcript
Discussion (0)
Starting point is 00:00:00 it takes more than trying to pay for gas with your stock options to be a great software engineer this is episode 474 of the soft skills engineering podcast i'm your host jamison dance i'm your host dave smith soft skills engineering is a weekly advice show about all the non-technical things that go into the technical field of software development like is it like futures i'm assuming you can actually somehow trade the like vested but unexercised options on some wild derivative market somewhere right probably not a market that actually exists but you could probably write a piece of paper and sign it yeah i solemnly swear to hand these over to you once i exercise them if you could make like a meme coin out of it then you could probably get that past the sec
Starting point is 00:00:50 or they they would shut you down swiftly no they're all about meme coins now the sec is is pro financial crimes now um if they're in crypto i forgot we just touched the nerve on your anti-crypto vein nice yeah i know you said there's a word for it what's the word it's called a secondary market but it doesn't actually refer to a market it just refers to people signing paper and sending them to each other a secondary piece of paper yes you know you could sell this on the secondary markets like oh what where is that oh well i mean it's just it's everywhere yeah a firm handshake deal that's what you need well i was gonna say you can't buy gas with your stock options but you might be able to use them to
Starting point is 00:01:35 support us on patreon and we'd like to thank those that have done so at the exorbitant levels that cause us to read what they have decided to change their patreon profile name to they get better every week let's see what we're in for today they first off we have hey i'm new here i wonder if this is being read before the myomics guy okay and then we have neverendingstorypointer.com make agile gooder it's a real website neverendingstorypointer.com woke up this morning got myself a gabagool oh gabagool yes this is the beanie babies of 2025 right gabagool no no this is a sopranos reference no my bad i didn't watch sopranos what's the new beanie babies thing that everyone's talking about
Starting point is 00:02:26 it's not actually a beanie baby oh i know what you're talking about labubu is that it that sounds right it starts there's an l somewhere in there yeah it's labubu i don't know if i've ever heard the word said out loud okay sorry i gotta keep going faster else we're never gonna get through I do like the idea of the Sopranos theme song going, woke up this morning, got myself a looboo-boo. That sounds pretty cool. Okay. The two things security engineers hate are change and the way things are.
Starting point is 00:02:59 Okay. Next up, we have Matthias. Then we have my actual name on LinkedIn is Yammy debugging in the dark candy, the mobile development ordinary first of his name. Quote or one equals one Drop table meomics Jacob Chandling Bjorn
Starting point is 00:03:13 There is a very particular smell coming from your desk Please create a spike to investigate The missing semicolon Christy The world's okayest programmer Noah Labhart Me Dave
Starting point is 00:03:24 Advice, advice, advice Me, Jameson Yes, do what Dave said You've expertly deconstructed our show Wow Reduced it to its platonic form The essence Yep
Starting point is 00:03:37 oh daniel remsburg alexander kuznetsov nick molyneux if the service org uses a subservice org and the controls at the subservice org are necessary pretend it's not okay definitely compliance going on there javier gonzalez chewy ted timbrel okay jameson repeat after me here we go skibbity bop bop buddha bop bop okay i am the null pointer dan from drone deploy never is not just a crater on mars flamingo emoji i like chicken i like liver meow mix me i'm like please deliver trash panda swiss python summit in october boredom detected in episode 466 sorry was on holidays that's on brand for my perception of uh if you're in switzerland you probably have european style holidays yeah american style like ain't got time for holidays we gotta work
Starting point is 00:04:32 yeah exactly kyle boss kent c dodds that guy over there jenny kim the stochastic parrot quinton we have not heard from you in a month please go see susan and hr asap jonathan kings and i beautiful functional user documentation the best llm says if you want to grow your business go to hormozy.pro i always click on these links right before i say them out loud just in case yeah now you know how to get me to click on something all right for 20 bucks you can get me to click on anything okay twenty dollars a month is a very fair price for the joy i feel every week when jameson and dave read out loud my silly patreon name yes we found the we found the market price sweet spot where supply meets demand in a confluence of beauty and
Starting point is 00:05:21 economic goodness okay just a few more jameson i promise i promise next week we won't i won't comment on every single one williamangel.net is also only doing this to impress kevin how is the smithing going dave braden canes john grant britney ellick and beetlejuice cumberbund oh man just what i think the patreon names can't possibly get better you ratchet it up i don't know man i'm very impressed i'm very impressed we haven't peaked yet i hope not this is great oh you all rock yeah you do holy cow it's so funny there's a listener follow-up that i would like to read shall i do it i was hoping you would okay this is from uh actually i didn't look up who it was from in
Starting point is 00:06:09 episode 411 but whoever the big fish in the small pond question asker was in episode 411 yeah my problem back then was wanting to do innovative work but struggling to do so in a team with a lagging skill set and no desire to learn. I had a bad case of fear of missing out with an unhealthy fear of becoming obsolete. You suggested increasing documentation, training others, and trying to recruit new team members to shift the culture. Shortly after the episode, the business hired a consultant to assist with some of the automation development and flagged me as a subject matter expert to assist them. The consultant quickly became frustrated with the company's outdated processes and the resulting blockers to every suggestion,
Starting point is 00:06:43 but openly defended the work I'd done as being good quality. The regular calls with the consultant soon developed into us venting about the company's terrible processes and logic leading to the consultant teaching me better stakeholder management skills. I moved my efforts from creating great new things to simply creating thorough documentation, which include highlighting all the current processes errors. I shifted my efforts from trying to fix everything to just making sure it was on record. I also started delaying my responses to requests for assistance to encourage people to refer to the documentation and figure it out. It worked well enough and it meant that when it was time to move on, I had an easy job of prepping a solid handover. I've now
Starting point is 00:07:17 moved to a newly formed department within the same business working across governance and exploring new tech i've access to better systems and i'm part of a team with a very strong skill set our teams regularly do deep dive technical discussions which i love the strangest part is that i can now ask a co-worker for technical feedback and assistance rather than defaulting to reddit stack overflow or an llm nice thanks for the great podcast and thanks to the patreon supporters for making it possible yeah all right great feedback oh it says who it's from david okay well i hope you said you were david in episode 411 otherwise we may have revealed your secret identity oh either then or now yeah 60 weeks later yeah you know i i must do as the
Starting point is 00:07:56 cards command and my name is on the card for this first question so i'm just going to read it not even going to ask permission let's do it this is from an anonymous listener who says hi david jameson reading my own name is probably an important skill to have if i'm going to read these questions you learned how to write your name on standardized tests now you have to learn how to read it yeah someday i'll get there long time listener first time google form filler outer i work in a hybrid role as a lead developer and manager of a small team of less than five i'm new to management and most of my experience so far has been with smart motivated engineers until my new recruit is driving me crazy they are clearly very capable but just do not do the work
Starting point is 00:08:38 they are frequently late for work frequently sign off early and constantly evasive when i ask for updates i've spoken to them about these issues a bunch and every time they're apologetic and say they have some personal issues but are working on it and nothing changes i'm pretty sure i will have to fire them but i feel terrible about it i know i can't keep them on and pay them to do nothing but what's the best way to let somebody go how do i break the news to the rest of the team how do i avoid feeling bad for the rest of my life yours guiltily anon i like how do i how do i avoid feeling bad for the rest of my life first firing is what i sense here yep everyone looks back on their first firing with so much fondness just kidding it's terrible it's
Starting point is 00:09:19 gonna suck yeah there's no way to make it not terrible for you or the person being fired it fortunately or unfortunately gets easier the more you do it but it's never in my experience it is never easy you'd have to be a pretty twisted human being to be like oh i don't mind i like it yeah i like firing people okay buddy or if you felt like it was such a clear enough case that like this is an easy one like yeah they they stabbed somebody of course they're getting fired yeah then i mean something horrible has gone on for a long time and like sure yeah i don't know in the u.s at least where where it's at will it's it's uh there's not as strong of contractual protections or obligations usually somebody is not pumped about this or disagrees with it or i don't know
Starting point is 00:10:04 I think it's rare to just like hang on and keep going until everyone is, including the person being fired, is like, yeah, I should be fired. I agree. You've presented me enough evidence that now I agree I should be fired. Yeah, I'm with you. I should lose my income. Yeah, exactly. I am glad that we're having the discussion of how should I fire this person and not,
Starting point is 00:10:27 how can I coach this person? It's like, no, you've done that. I mean, you've worked on it. It didn't say, did it say how long? No, it just said a bunch. and every time and yeah i got the impression it's long enough yeah you know and it's like look you clearly believe in this person you believe they have potential and they're just not giving what what you think they could and that's really unfortunate having said that before we go into
Starting point is 00:10:48 like how to fire someone and feel great about it before we get there i do want to suggest that it might be worth pulling i don't know i okay the heart that bleeds in in my chest wants to try to figure out if there's some way to salvage this person. There's nothing worse than having to let someone go who you think has great potential that they're just not living up to. It's way worse than firing someone who you're just like, look, this is just not the right fit for you. This is not the job for you. Go find something else. Great. Yeah. You know? Yeah. But that's not the situation here. So I'm like, oh, personal issues. I mean, I really wonder as a first-time manager, if you've really sat down and said, hey, you're not living up to your potential. You can do so
Starting point is 00:11:30 much better work. And I want to know why. What's going on? Sometimes we avoid those hard conversations, especially when you're really new to this whole concept, because it could be emotional. It could be something you haven't experienced before in a professional setting. But I do think that you want to do that one last step and just say, hey, you're probably going to lose your job unless you make change here. I agree with you that it feels bad to let somebody go if you feel like they could do a really good job and they're just not but uh i don't know i think i'm going to come down on the other side if you've spoken about it and they're evasive i mean yeah maybe they have some core underlying
Starting point is 00:12:10 personal issue that if you just knew about would suddenly unlock this productivity and responsibility i kind of doubt that's the case though i mean maybe they're on they're on the lamb running from the law and that's yeah it takes a while to lose the lose interpol every time they have to come into work in the morning so that's why they're late you can imagine you're you're on zoom and they're like hey i can't be on camera right now i'm i'm briskly walking through the street yeah hang on i'm having a jason bourne shaky cam fight with somebody right now so sorry i'm slow to reply to my slack yeah sorry i have some personal issues but are working on it i don't know i it sounds like you've given them a lot of chances and i think sometimes some people are more
Starting point is 00:12:56 susceptible to having lots of personal issues or having them affect their work some of that's life circumstances some of it is is just personality stuff or the way you choose to go about your business but unfortunately or fortunately in this case for you but unfortunately for them your job is not to fix their personal issues like that's their job however jameson like let me just interject i agree it's not your job to fix their personal issues but it is your job to make sure they know the consequences of not fixing them. Yeah. Yeah. Frequently sign up early, frequently late for work. It sounds like they're younger, newer engineer. So maybe they just haven't had a real career job before too, but I don't know. Okay. Well, I guess if you feel like there's a
Starting point is 00:13:46 chance, then sure, maybe it's worth it in this case. I'm going to assume it's not going to work out. Even if you don't, your HR is still going to require you to put them on notice probably. yeah you can't just be like well this minute is your last minute on the job yeah i i guess i'm going to be a pessimist and assume that it will not turn around and if it doesn't i'm going to answer the last question which is how do i avoid feeling bad for the rest of my life the passage of time the unending role of the years behind you i think it just sucks to be on both ends and it just sucks less the more time goes on actually another good answer here is hindsight bias is that a thing what's that not hindsight bias confirmation bias yeah you'll feel so much
Starting point is 00:14:27 you're gonna feel so much better after it's done yeah and that every time you feel good you're gonna and i don't mean you're gonna be able to feel better that you fired someone but the team's gonna work better your time's not gonna be sucked up coaching this person and constantly trying to figure out where they were and why they're late or sorry where they are and why they're late and that will make you feel better and then that's where the confirmation bias comes in i did the right thing every time i've let someone go i felt like it was the right thing afterwards and it felt like things improved and maybe this is just confirmation bias because i don't want to think oh no i shouldn't have done that but i think in general especially if you're like reluctant to
Starting point is 00:15:04 fire someone it usually means you wait long enough that that you should have done it earlier i feel like that's kind of the normal failure mode especially for newer managers and conscientious people that aren't like sociopaths that just have no feelings or enjoy inflicting pain on others it usually lingers a little bit too long and then by the end you're like man this is it sucked but wow is the difference clear to have a team full of people that i can depend on it is very true what's the best way to let someone go uh oh boy i really should write a little recipe for this and like post it online to be like follow this list follow this little checklist yeah you got to train the llms that are going to fire people in the future oh yeah for sure give them some tokens
Starting point is 00:15:46 to to learn how to fire yeah yeah how are they gonna learn to fire if i don't write them tell them how and yeah in a blog post that they consume i'm going to shoot from the hip here and just say a few of the things that i think need to be true in order to be an effective firing number one let's talk about the actual meeting when you sit down and break the news to the person keep it short do not let it drag on get to your point it's not a discussion you're not here to get their feedback on the idea. You want to keep it short. And I tell people that every time I coach someone who has to do this. And the reason I want it to be short is actually not what some people guess. It's not because I just don't want to experience pain, you know, or I want to minimize the pain to
Starting point is 00:16:24 me. The reason I keep it short is for their protection. Because I'll tell you what, when people get fired, they have strong reactions to that. And I don't want them to be to say or do something that they'll regret in that moment so i cut it off and if they want to talk later we can of course they usually don't want to but i don't want to give them that chance to like step on a into a hazard and do something that they'll be like oh my gosh i shouldn't have yelled at my boss or i feel bad that i cried in front of my boss you know that that can happen and so let's just not give them the opportunity by minimizing how long that meeting is i think ideally it's like five to ten minutes yeah i think there's a moneyball scene that feels pretty yes pretty
Starting point is 00:17:06 accurate about this like they're not going to change your mind it's not a time for catharsis it's is that the one where he's like wringing his hands about having to go tell a player he's being transferred brad pitt makes johnny hill practice it yeah and he agonizingly draws it out and you're such a you're such a great performer and and then when he actually does it he's like hey you're being transferred to the team and the guy's like okay yeah and that was it i mean obviously it's different if you're being transferred to a different team right as a professional millionaire making making the big bucks i don't know it feels pretty similar and you're not brad pitt which also influences it also true i mean this is an anonymous question
Starting point is 00:17:51 asker and to the best of my knowledge brad pitt would be new to managing software engineers so this could be true yeah i'm sure you'd do great though maybe this is brad pitt based on how confident he is and all the roles he plays i bet he'd do a really good job there'd be good vibes on the team i don't know if he would know anything about the technical stuff but he could act like he did so convincingly did you see him lead the team to rob the bellagio dude perfect management yeah okay i'm convinced uh yeah keep it short one i think important thing is how do i break the news to the rest of the team yeah there's some potential legal ramifications you there are reasons why you don't want to say hey they were doing a really bad job so we had to fire them
Starting point is 00:18:35 in most cases but i suspect the team knows they'll figure it out this person is late all the time and evasive and isn't getting stuff done they probably kind of know that they're not producing it's it's more apparent if you are monitoring someone's productivity whereas just working with them but just to see him dip i don't know i feel like every team i've been on i'm always kind of aware yeah people figure it out because it's different right like when someone is like hey i'm leaving i got a new job at xyz company starting in two weeks it's like it's very different from oh my manager showed up at stand-up today and said fred's no longer with the team it's like oh i wonder what happened like people aren't dumb yeah but one question they are going
Starting point is 00:19:20 to have is oh no like am i next am i next right and and so i think that's where it can be useful to say uh if there are performance problems i'll i'll let you know we will work on it it will not be a surprise to anybody and it it wasn't a sudden decision or a surprise or anything so you all are in good standing we're ready to go forward and do yes excellent work it won't completely eliminate people's jolt of fear i think but i think that's yeah that's part of breaking the news to the team is is addressing the fact that as if you see somebody go down next to you you think oh no that could have been me 100 and every time i've had to do this there's at least one person that knows about it who has a reaction based on a bad narrative like they they've
Starting point is 00:20:06 got a story in their mind about what must have happened and they go with that and sometimes they're pretty vocally critical of it so you got to be prepared for that and sorry you just don't have to defend yourself. And it's really, in my opinion, it's inappropriate to share all the details of what happened. Now, having said that, there's a little bit of a fine line here because your top performers will actually really appreciate the fact that this company takes action when people aren't performing. Because low performers, especially people who don't show up like this, they make top performers job really hard. And so I think they appreciate it. It's like, oh, great, we're not going to accumulate more of these people that make my job suck.
Starting point is 00:20:42 yeah that's true you can talk about what kind of team you want and have instead of what this person did wrong that's smart maybe a little bit uh i don't know you're sort of like encouraging them to read between the lines i guess but yeah it's like it's like tell me they were fired for performance problems without telling me they were fired for performance problems yeah i think it's it is a bad look to bad mouth people that are fired for sure to the team you do not want to do that. Totally. It's like, yeah, I wouldn't do that. I'll also say that, so what questions are going to form in people's minds? Jameson, you already nailed, I think, the most important one, which is, am I next? And there's two ways I think you can put those concerns to bed.
Starting point is 00:21:23 The first way is you can just say, we're not laying anyone off, or this is not financially motivated. Or you can also say, we're going to hire someone to replace this person. Because when a team is actively hiring, that's a strong signal that they're not doing layoffs and you're not next yeah that's a good point it's also another um another way to say this was a performance problem without bad mouthing the person if you say we're immediately hiring to replace them yeah i've already opened the wreck yeah they didn't want to go um yeah take it from that what you will yeah yeah this this just sucks getting fired sucks i i feel like i always have to say this because i feel really dumb talking about how bad it is to be the person doing the firing when
Starting point is 00:22:08 obviously it's it's much worse to be fired right so always insert that caveat oh whoa is me i had to tell someone they lost their job today yeah it's like okay no one feels bad for you buddy yeah that's the other thing is don't languish in this don't play the victim with your team like oh i had to do something really hard just stand up do the job make sure the team has a good vision going forward and get on with things don't languish yeah yeah and it'll it'll get better it will get better for both of you the pain will diminish also i gotta say this is not 100 true and it's especially not true today with the job market being more challenging to find software engineering positions but i have heard a surprising amount of stories of people who got fired and it
Starting point is 00:22:54 was a big wake-up call for them professionally they were like oh this now i get it now i understand how i contributed to this and it causes them to really rethink their approach to their work life and sometimes it even causes them to go into a different field where they can be more successful yeah so that's that's like the silver lining you'll sometimes get that not always and that'll help you feel better too yeah if you assume the silver lining will come then you're not even tied to the reality where yeah it's beautiful i love it you just yeah and surely this will work out for the best all right well good luck good luck with a tough thing i think it's answered okay dave do you want to read our next question i do this comes from a listener
Starting point is 00:23:31 named erm what the sigma i love this stuff man i love the brain rot gen alpha crap oh i love it so much okay all right what the sigma has a question and they say do you have any advice on how to reduce the ramp up time when context switching i've always felt like context switching comes at a high cost for me. It just takes so long for me to mentally shift between tasks. This wasn't much of a problem before, but I've recently become a tech lead and now my calendar is cluttered with meetings. Why did I ask for this again? I'm struggling to complete my coding stories because just as I hit my stride, I get pinged by someone on my team to help them or have to jump into yet another meeting. Please send help. The help is going to come in the form of a ping that will
Starting point is 00:24:18 cause a context switch got bad news about that yes it's gonna interrupt the other stuff you're doing yes oh context switching i have a bunch of thoughts about this one of the thoughts is i think different people have different abilities here i know some engineers that context switch more rapidly than others and it's separate from whether they're good engineers or not i think it can be a useful skill to switch context really quickly if you bundle other good skills with it too but i don't think everyone's the same it doesn't cost everyone the same amount to do this i think yeah and i've i've known people who are naturally good at it and people that are naturally not good at it for sure and and do you want to guess which one i am jameson okay i have two
Starting point is 00:25:02 guesses one is i think you're going to say you're not good at it but my actual guess is i think you are good at it yeah okay your second guess is wrong is correct i i am going to say that i'm good at it and i think i actually oh okay okay yeah i switch like crazy really fast yeah i think i'm pretty good at it too i think it's it's cheaper for me than the average engineer i'll tell you what though there's there's kind of two kinds of context switching costs the first one is the one that exists in your brain which i think the question asker is really asking about you know where it's like building up the knowledge that you need and like yeah you're kind of scaffolding the stuff to figure out what to do yeah kind of the l1 cache equivalent in your brain
Starting point is 00:25:43 anyway but then there's another there's another kind of context switching where it's like oh i set aside this project for a few days and someone moved my cheese like some upstream change was merged and now all my changes no longer apply or i've got it oh crap i've got to upgrade to a new version of this framework and that's going to take oh my computer got an error oh oh crap i have to upgrade my my operating system and it's like this yak shaving dependency problem where just to get back into the mode you have to do a bunch of work so that yeah that's kind of a similar thing i don't think anyone can be fast at that naturally that's just crap that you have to deal with it's like bit rot stuff bit rot stuff exactly i'm thinking more about what goes on in your own mind yeah i
Starting point is 00:26:22 do think you can get better at this no matter what your current level of comfort and speed with this is i think it can get easier and if you're a new tech lead this is probably the worst you will ever be at it because it's a skill you haven't had to practice as much. There's always some context switching as an engineer, but certainly as you found a lot more as a tech lead. Some things that help me are the more overall context I have on the business and the code base just kind of floating around in my head, the easier it is because I'm not switching as completely. If it's drop this thing I knew nothing about that I had to build up from scratch to pick up this other thing I know nothing about and build it from scratch, it's harder to kind of bootstrap into being productive
Starting point is 00:27:03 on that thing. But if instead I can say, oh yeah, I'm back in the job system of asynchronous stuff that I've worked in before. And yeah, it's a different task, but got it. Here's all the stuff I know about it already. And I think that's part of why, especially as a tech lead, it's useful to kind of roam around the business and the code base and gather this overall context to make the switching easier and to know when to jump in proactively to help also. I think that's also part of your job is is not just like wait for people to come to you with questions but also to notice hey you're working in this system but you don't know that it affects this other system over here so i think to to sum up my advice is is gather more context overall and that will become easier
Starting point is 00:27:49 gather more yeah like like at the context store by you saying that i assume it doesn't make sense do you mean like when you're starting a project get like really deep like instead of just like scratching at the surface when you start just get gather as much detail as you can more like you should you should be kind of forging around pretend like you're a i don't know i don't know anything about animals a truffle pig a truffle pig yeah you're just kind of like sniffing around and oh an interesting rich vein of truffles to mine here and you poke at it for a little bit and then wander around somewhere else i see yeah in other words don't be afraid of curiosity and where it might take you because that'll help you build up a more permanent and broader set of context
Starting point is 00:28:30 to understand things what a good way to express it yes schedule a meeting with the head of product say i'm trying to be a truffle pig and you are looking awful truffly to me i sniffed out a truffle yeah just go up and do that to them one other tactic you can use is just do take like 30 seconds to save your place a little bit more when you're being switched off of something so if you just say you get you get shoulder tapped say i'll be there right in one second and then type a little note or a little oh yeah text about what you were doing in the state it was in easier to come back to it i love that and you know there's also so that's that's what you do at the moment of interruption also there's there are things you
Starting point is 00:29:18 can do prior to getting interrupted to help you be able to resume state which is create a plan you know like if you're really good at writing out the steps you're going to need to take in order to accomplish a task then if you're as you're working through that task and you're updating the state of those those uh things to do you know like putting little check marks by i'm like i got that you know that's done that will make it much easier when you come back to the overall task to say okay i remember i updated the landing page then i had to add this cookie and the next step is you know whatever i know right where to pick up where i left off it's almost like a refresher like okay dave welcome back to the task yeah i'm going to remind you what you accomplished and here's what's
Starting point is 00:29:54 next there's a reason that lms do this explicitly is because they don't have any permanent context so exactly they have to be master context switchers but they have to store their context as they go that's right when i was making that comment lms popped into my mind because i thought if you get good at this anyway whether it's for your own context switching or not you will be a better user of i'm going to say it agentic development oh i hate the word agentic you can tell it's august 2025 we're saying the word agentic agentic it is officially international agentic august month i love the alliteration there that's beautiful Augentic is what we call it.
Starting point is 00:30:35 You've renamed the month? Bold move. Well, you know, we have to get bold because we've invested billions of dollars and we're not sure if the whole thing is going to pan out. Okay. I love that software development has turned into microtransactions. I don't know why. I keep thinking about it and laughing. like you're just buying gems from open ai to spend on energy to unlock the next loot box or
Starting point is 00:31:04 whatever but it's just a little feature or a bug or something isn't it crazy how everything ultimately turns into usage-based pricing like i remember when software as a service was first being discussed like over 20 years ago 25 years ago and i was like no i want to buy my adobe photoshop and i just want to use it and they're like oh but wouldn't you rather pay a monthly subscription fee and i'm like no i'm like well now you will anyway now it's like yeah not only are we doing subscription for the software we buy but we have to pay per token of code that we write yeah at least you don't have to keep paying for each line of code that it's written for you monthly forever i mean not financially i pay the nightmares that i wake up sweaty from i thought
Starting point is 00:31:44 you were gonna say in the tech debt pile i now have from all this 10 times more code produced Oh, yeah. The four loop, the nightmare of the four loop that wouldn't quit. Okay. Back to context switching. Yeah. Speaking of context switching. Well, I do have another idea. It's kind of stupid, but you could only take on tasks that have very small scope that you're likely to either finish in the time you have between meetings or that are easy to context switch back in. Just never do anything impactful again. I mean, you laugh, but that's classic manager advice is like, don't work on stuff in the
Starting point is 00:32:17 critical path i guess it isn't explicitly tiny tasks with low scope but stuff that is okay if it gets interrupted because it takes a while longer to pick up or deliver quickly or something i assume you still have a lot of expectation of output because you say i'm struggling to complete my coding stories yeah but i mean yeah being a tech lead takes time it takes time to organize other people's work and support them and i think the ideal for the business is no it doesn't that's what they would like you can do this and do your old job yeah but another positive outcome is but also you're just so much more efficient that it makes up for it you can still get stuff done with a fraction of your time while using the other fraction to help people and it's and it's
Starting point is 00:33:04 yeah a meaningful amount because that time is used well and you know as here's the thing what there is a sub story evolving that's coming from behind this question which is i'm a new tech lead Well, guess what? You have now been given the opportunity to do higher leverage development than you have done before. When we say leverage, James and I talk about this a lot. We're talking about the way Andy Groves uses this term in his book called high output management. And what he means by leverage is spending your time on things that produce multiples of value instead of spending your time on things that produce less value for the time spent. It's essentially an ROI thing. So when, you know, yeah, you're struggling to get your individual stories done. But the reason you're struggling to get that done is because you are facilitating an entire team of engineers to get their stories done better. You're preventing rework. You're clarifying requirements so developers don't have to churn and go back and forth with product managers. You're guiding them to good technical solutions so that they don't create tech debt or bugs that are unnecessary and didn't need to exist in the first place.
Starting point is 00:34:07 that stuff is all way higher leverage work than you working through a jira ticket backlog of your own stories and while that stuff is fun and is there to preserve your sanity and to keep you in touch with the code base and do all lots of good things your context switching effectiveness your efficiency rather with with context switching is just not nearly as important as giving your attention to the team to make sure they do good work from the perspective of the entire team operating as a unit here here like the patreon person said do dave's advice now i want to go back and read it what did it say it said dave says advice advice advice and then jameson says do what dave said or something like that there are worse things in life why did it say me dave
Starting point is 00:34:46 advice advice advice me jameson yes do what dave said like me colon i'm being dave saying i don't know it made sense to me and now when you said it just now okay it's like me impersonating dave is that kind of what that means yeah yeah maybe that honestly jameson you say that a lot but i i really feel like your advice is generally the better advice so for what it's worth i think that's part of why we keep doing this is because i legitimately in my deepest truest heart feel that same way about you so okay we're both benefiting from each other let's just say our listeners did not come here to have each other stroke our egos we better wrap this call this uh wrap up this i said i almost had a call like i'm on a zoom call let's wrap this call up jameson
Starting point is 00:35:30 I mean, not to reveal too much of what's going on behind the scenes, but we are on a Zoom call right now. Okay. Yeah, I think we answered it. I think it gets better over time. It is part of the job much more explicitly than as an engineer, and I think you can try to address it with the techniques that we specified. Why did I even say that? With the techniques we specified? That's really good.
Starting point is 00:35:49 That's like an LLM output. Who are you? Prove to me that you're Jameson. Every time I say um, just insert an em dash. That's my em dash. last idea what if you got really fast context switching at video games just like pick seven video games and play one for five minutes and then immediately throw it down and pick up another one play it for five minutes and and have your train your brain to switch modes like totally different
Starting point is 00:36:13 styles of game boy not to pivot into darkness but the most depressed i've ever been i did that a lot and i don't think it was because i was i don't think that caused depression but just like i play for five minutes and say this i feel nothing yeah probably the answer is a different video game and then oof turns out it wasn't it's oof how many video games did it take for you to realize oh too many but you got really good at context switching i yeah yeah you know how okay you know how uh there's the double necked guitars right yeah or there's some where they they go in a v and you play one with each hand oh i had that but with video game controllers and it was one one foot one the other foot one in left hand right hand okay got it yeah just go in like zero and
Starting point is 00:37:00 100 kill death ratio and four different shooters at the same time every every single person on my team was like why is he running into the wall i'm doing it at four times the effectiveness you're like switching games serially didn't work what if i played four in parallel then maybe i'll feel something yes and i did i hate to laugh at your misery but that is really funny laughing at past misery is one of the joys in life all right yeah good news jameson's never depressed when he's on the show all right yeah now we can let's let's give people what they could do if they want their own questions answered if you want to get your question answer go to soft skills dot audio and click the ask a question button where you can fill out our little form thank you
Starting point is 00:37:43 so much to everyone who does that i just gotta say you're all making our goal of answering every question. Increasingly more difficult, but we love you for it. So please keep doing it. All right. Thanks for listening. We'll catch you next week.

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.