Derrick Tebusweke (@DTebusweke) header
Derrick Tebusweke (@DTebusweke) profile picture

Electrical and Space Systems Engineer | @kyutech and @Makerere alumnus | Building @cansatug

Earth
Joined August 2010
Photo by Derrick Tebusweke (@DTebusweke)Photo by Derrick Tebusweke (@DTebusweke)Photo by Derrick Tebusweke (@DTebusweke)Photo by Derrick Tebusweke (@DTebusweke)Photo by Derrick Tebusweke (@DTebusweke)Photo by Derrick Tebusweke (@DTebusweke)Photo by Derrick Tebusweke (@DTebusweke)Photo by Derrick Tebusweke (@DTebusweke)Photo by Derrick Tebusweke (@DTebusweke)Photo by Derrick Tebusweke (@DTebusweke)
Weekend read.
6
3,918
Closed mouths don't get fed. I was simply requesting for a chance to serve my country's booming space sector. Either way my first degree can keep me busy. I know the powers that be recieved my plea. Aluta continua
You were begging for jobs recently now you’re yapping online stay where you are don’t pander
1
62
As a Master’s degree student, ensure you publish your thesis in a Q1 journal. This will help you later on when you're searching for PhD Scholarships as most have that requirement. Your Professor might not even tell you this, but it's very important. Even so for Bachelor's degree.
5
105
Derrick Tebusweke retweeted
读博以后才发现,科研能力的差距,很多时候不是智商拉开的,而是一些不起眼的小习惯。 下面这6个习惯不复杂,但坚持一个月,确实能少踩很多坑: 1⃣每天固定时间看文献。 不是等写论文了才突击下载几十篇,而是持续关注本领域的新论文、新问题和新方法。 优先看预印本、顶会论文和目标期刊的最新文章,哪怕每天只认真看一篇摘要,也比月底集中补课有效。 2⃣用AI筛选和整理文献。 先让AI根据标题和摘要判断相关性,再提取研究问题、方法、数据、结论和局限,最后按主题归档。 但别让它替你判断论文价值,关键文章必须自己读原文,尤其要检查实验设置和作者真正得出的结论。 3⃣每天写实验日志。 今天改了什么参数、为什么这样改、出现了什么异常、哪些方法失败了,都随手记下来。 很多坑当时觉得自己肯定不会忘,过两周再看代码,连当初为什么删掉一行都想不起来。 4⃣有想法就尽快做最小验证。 不要一上来就搭完整系统、跑最大数据集,先用最简单的数据和最小实验验证核心假设。 结果不行就早点换方向,别花三个月把一个错误想法做得特别完整。 5⃣多和别人交流。 把实验思路讲给师兄师姐或不同方向的人听,很多问题在开口解释的过程中自己就能发现。 AI可以陪你推演,但它经常顺着你的错误前提继续编,真实的人会直接问你:“这个假设凭什么成立?” 5⃣辩证使用AI。 AI可以帮你找资料、改代码、整理结果,却不能代替你理解实验的底层逻辑。 至少要说清楚输入是什么、变量为什么这样设、指标代表什么、结果能支持多大的结论,否则代码跑通了,也不代表科研做对了。 我自己最想提醒的一点是:别只记录成功结果,失败实验反而更值钱。 它能告诉你哪些路已经走过、为什么走不通,也能防止半年后自己又把同一个坑踩一遍。 科研进步很多时候不是突然想出一个SOTA,而是逐渐学会不再重复犯错。
对 PhD 一年级新生有什么建议? 1. 只为自己工作,这跟付我钱的导师,其他人都无关。 2. 无论想不想以后走学术路线,都以做faculty作为职业目标。尽管出成果需要运气,但是以faculty作为目标,结果都不会太差,会收获不一样的人生。 3. 不要随便给自己放假,常年持续性工作。每天如果能工作12个小时,尽量坚持到14小时,因为自己的所有产出,可能都是最后两个小时做出来的。这个建议对一年级新生尤为重要。 4. 博士生在哪个国家永远是身份最低的存在,跟博士不同,千万不要有优越感。博士生是一个个体水平差异巨大的人群,在你不知道的学校,可能你的水平都会被你同行的本科生和硕士生完爆。我们永远需要极其努力工作来impress周围的community,来获得更多的资源和更高的平台。 5. 同样的实验结果,present的方式不一样,取得的效果和为成长铺路的功效天差地别。这适用于group meeting,paper,proposal,谈合作,几乎贯穿科研始终。 6. 一切以把事情做成为中心,自己的情绪是最不重要的,不要太脆弱,把自己太当回事,绝大多数没人会在意你的情绪。最开始的时候,自己的可替代性非常高,项目不是少了自己就不行。 7. 追求独立科研,不要觉得应该跟哪个师兄或者博后老哥干活混。没有好下场的。是否该和跟博后或者senior同门合作有两个标准: 第一,这个人确实有很强的一作工作,非常有insight。(可以跟其学习) 第二,这个人为人非常nice,会为人着想 (不会坑你)。这两点缺一不可。最怕的就是空有一把年纪,学术水平却跟不上,心里都是小99,把小学弟学妹当傻子。这种很多,尽早脱离为好。 反过来,等自己成了高年级学长,如果觉得学弟学妹值得自己的时间,一定要好好培养新人,对其负责。 8. 遇到任何人,笑着打招呼,善于称赞别人的工作,保持良好的人际关系。 9. 没有任何人做工作(导师,同门,同学)会首先为你着想,你也同样。但是在喂饱自己的过程中做到顾全大局,try to make other people successful if he/she worth your time。如果你有货币(能力,人,技术,资源),该花就尽管花给别人。 10. 高端科研圈里,如果你觉得这个人(导师,牛逼师兄)很高尚,不图这不图那,但是事实上往往不是这样的,他想要的比你想象的多得多,只不过那些东西跟你想要的不是一个level的。 小萌新只想发点小paper赶紧吃到点小甜头,人家满脑子只想发nature science,拿几百上千万的经费。他们眼里,只有自己头顶一片天,其他都看不见。反之,如果你发现一个长辈还在跟你斤斤计较一些小事,也可以果断离他远一点了。时间宝贵,与其接触降低自己level,徒增烦恼的事少做。 11. 可能的话,多吃小亏。没人是傻子,如果有一次耍小聪明还当别人是傻子看不懂。你在很多明白人的心里就直接被pass掉,亏大了。 12. 跟人约meeting,无论是干什么,都要严格守时。不尊重别人时间是大罪过。如果有约别人迟到,我只可以忍一次,第二次第三次之后,果断断绝一切接触。 13. 研究组或者高校环境类似动物园,强者可以拿到所有风光无限,弱者蜷缩在角落被人宰割,其疾其苦nobody cares 14. 往往只能吸引到跟自己same leve的人做你的collaborator,所以,坚持不懈提升自己,但是努力跟比自己更强的人打交道,与其一起学习进步
2
174
7
1,617
99,104
Hey @MODVA_UPDF , the US Army's enlisting age was updated up to 42 years old, why aren't you recruiting past 34 years? The world has changed and many 35+ are likely more educated, fitter and willing to serve.
1
2
135
You don't employ 300 people amd you're still sleeping? 🤣
1
120
How does a country which had a 20 year war and who's GDP is 3x smaller than ours look better than ours? Is it possible to be poor and organised?
Russian Deputy Prime Minister Alexey Overchuk in Kabul. Highest level visit to Kabul in years as Russia, Afghanistan firm up ties.
1
94
Grown adults here fly-tip without an ounce of remorse.
??? I'm assuming that people who come with their fancy cars are in the top 10% of earners in the country, yet animal behaviour persists. We LITERALLY are NEVER making it.
68
If you check those drainage channels now - you'll find trash. Prove me wrong.
The Nigeria You Will Never See on TV 🇳🇬 Nigeria is the United States of Africa. Abuja should be expecting foreigners for detty December.
54
This is where I stop.
35
Use Consensus to gather papers on your topic. Then Zotero and Google LLM to filter map your found research papers. Then read only 20 to 25% of the top papers obtained to find gaps / concepts. Ask any LLM for a structure, then write your content. Finaly AI and Manual review.
Do not write your dissertation with AI. You are hurting yourself doing that.
1
1
178
We be busy chasing deadlines and forget that Uganda is beautiful.
2
1
4
224
Derrick Tebusweke retweeted
I am loving this side of X. Keep bringing this my way
𝗛𝗼𝘄 𝘁𝗼 𝗪𝗿𝗶𝘁𝗲 𝗬𝗼𝘂𝗿 𝗙𝗶𝗿𝘀𝘁 𝗣𝗮𝗽𝗲𝗿
4
213
1,938
44,184
Derrick Tebusweke retweeted
You don't need to know exactly where you'll end up. You just need to keep becoming more capable. Capability creates options, options create opportunities, and opportunities create futures that were invisible when you started.
4
141
5
1,095
16,613
Sharing for my PhDing network.
博士生写论文就要暴力产出! 现在找工作对于论文质量数量要求都很高,所以博士生第一要务就是快速写出论文: 1⃣️先找到一篇能“照着走”的目标期刊模板 别傻乎乎从头开始,先锁定一篇目标期刊里选题最接近的论文。不是学习观点,而是结构参照:每节干什么、哪里展开怎么收,讨论怎么呼应。只有吃透期刊风格,才可能入编辑眼。 而且这样写的时候脑子里始终有个终点形态,不会写着写着跑偏。很多博士写得慢,就是边写边还在反复想论文结构 2⃣️尽早搭建自己的文献与理论体系 科研不是囤文献,问题不在读多少,而是文献没有属于自己的位置。现在我整理文献或学习理论,就问一个问题:在论文里用来干嘛?是放在理论基础,还是方法合理性? 一旦文献与理论有了体系,写作时才不用翻来覆去找,而是直接调用 3⃣️学会高效使用AI工具 现在科研已经离不开AI了,但得先会选择,很多博士生都是同时打开很多app和网页,同一个问题到处问然后再比较,或者让每个工具都找材料,再交叉反复验证。这根本达不到理想效果,反而会拖慢进度 4⃣️把写论文拆解成有节奏、具体的动作 真正高效的人几乎不设“今天写论文”这种flag,而是直接写:改第二节第一段、补三条方法文献、压缩引言到800字。 目标一具体,心理负担会明显下降,只需要执行动作。很多博士写不动,其实是目标太大,启动成本太高 5⃣️拒绝论文完美主义 写得最快的阶段,恰恰是对初稿要求最低的时候。只要逻辑通、结构可行、别人能看懂,就往下推进。 反正后面一定会改,论文不是一次性成果,你一直写不出来其实就是完美主义作祟
1
187
Derrick Tebusweke retweeted
Okay, with practical skills and no money, how would one build a satellite or build anything anyway that is STEM related ?
I have always said it boldly and without any apology that people trained in STEM (Science, Technology, Engineering, and Mathematics) should not have to go looking for jobs. On the contrary, they should be the ones creating jobs. This is because STEM teaches practical skills. Skills that can help someone start a business, build something new, or solve a problem. So instead of waiting for someone to employ them, a person trained in STEM has what it takes to start something of their own and even employ others.
3
14
3
53
3,339
Replying to @DenisDukeUG
My son. You are a good journalist. Dig deep. There is a huge story behind this. He is shy to talk about it.
1
2
13
698
Dear HE @KagutaMuseveni I am one of the 3 Uganda engineers you sponsored to Japan to develop Uganda's first satellite (PearlAfricaSat-1)under @STIsecretariat but I currently have no job and would love to serve my country. My email: [email protected] Kind regards, Derrick
In exercise of the powers vested in me by Article 174(2) of the Constitution, I have named 13 persons for appointment to various government offices. I have directed the Public Service Commission to assess their suitability and advise me accordingly.
193
211
40
1,100
153,888
Derrick Tebusweke retweeted
On this day in 2022, Uganda officially joined the space age. Four years ago today, on November 7, 2022, Uganda launched its first indigenous satellite, PearlAfricaSat-1 into space before its successful deployment into low Earth orbit from the International Space Station on December 2, 2022. Led by Project Manager Omara Bonny and a dedicated team of Ugandan engineers trained in Japan through the BIRDS program at Kyushu Institute of Technology, this groundbreaking mission marked a historic leap for our nation's technological capabilities.
10
16
84
7,448
American space technology is protecting Lake Victoria🛰️ @FulbrightPrgrm Scholar Dr. Anthony Vodacek (@RITtigers) is training @Makerere students to monitor Lake Victoria's water quality with NASA satellite data, building skills to help Uganda manage its natural resources and strengthen water security. 🇺🇸🇺🇬
10
40
1
191
8,590
Derrick Tebusweke retweeted
One of the reasons I rate tall man(despite his naughty streak). He embodies the concept of you can just do things.
Another milestone in public transportation in Rwanda. A user can track when the next available bus will come.
8
99
2
361
9,457