HarshitDoshi

Love this! With integrations to 3rd-party data sources (like AirTable, Google Sheets, Databases, Notion, etc.), this can be a very useful product, especially for use with LLMs.

The trackpad calibration on my MacBook Air is almost okay, just a tad bit too slow/rigid to move around.

How about apps which only have the option of signing in with Google or some other social / 3rd party provider? Do you also not use other services like Microsoft's, Apple's, Facebook (Meta)'s? How do you in that case sign in?

Ask HN: For those without a Google account, why and how do you manage online?

5 HarshitDoshi 1 year ago 15 comments

Just started out building a micro-agency into software engineering and consulting. I left my 9-5 jobs in engineering and big-consulting to pursue my dream of building something myself. I love to code and build software and have always dreamed of bringing in change to the landscape around software engineering services.

https://shunyaek.se is my agency, "shunyaek"'s home-page. shunyaek refers to "shunya", meaning zero, and "ek", meaning one, in Hindi, my native language. It refers to the binary building blocks of the software & digital world. The landing-page is still under construction. Always open to feedback and crticism.

Résumé/CV: https://harshitpdoshi.online/resume/

Location: India

Remote: Yes

Willing to relocate: No

Technologies: Python (FastAPI/Flask), Javacript/TypeScript (ReactJS/NextJS/React-Router/ReduxJS/Zustand/Material-UI/TailwindCSS), Containers (Docker/Kubernetes/Podman/Dockerfile/Docker-Compose), Cloud (Azure/AWS/Terraform/Pulumi), Git/GitHub/GitLab/BitBucket

Email: [email protected]

How are the unique IDs created? And how do you manage collisions with such small IDs? I am not sure but I think the collision rate should somewhere be around "~1 day or 33K IDs needed, in order to have a 1% probability of at least one collision." as per https://zelark.github.io/nano-id-cc/ with an alphabet that goes like this "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz" and an ID length of 6 characters.

I have been looking for such a URL shortening solution for a client of mine and hop2 looks perfect with its minimal dashboard and simple usage. I am currently using a manual combination of Google Analytics and Bit.ly but it doesn't really work as I would. I have even tried building a URL shortener for this particular use-case and while I did succeed, adding analytics feels like a big task and an entirely seperate project of its own.

- Phase10.ToolKit (built a utility for playing the popular cards-game, Phase 10) (https://phase10.shunyaek.se)

SEEKING WORK

Remote: preferred

Location: India (Available in time-zones with partial overlap)

Full-Stack Software Engineer:

- Back-End: Python + FastAPI/Django, can work on data and analytics tools like NumPy, Pandas, OpenPyXL, PyPDF etc.

- Front-End: JavaScript/TypeScript + ReactJS/NextJS + ReduxJS

- Database: PostgreSQL, MongoDB, Redis, MySQL

- UI: Mantine, Material-UI, TailwindCSS, other niche ones too

- Other Technologies: Docker (+ Compose), Supabase

Latest Projects:

- Built an online suite of tools for a cyber-security consulting firm with features like scanning of internet assets like IP, e-mail etc, analytics using tables and charts, security self-assessment wizards driven through a questions-and-answers format, etc. Technologies: Python, FastAPI, PostgreSQL, Redis, Celery, JavaScript/TypeScript, NextJS, Mantine

- href•sbs (built a minimal URL shortener with focus on writing the CSS from scratch, without any CSS frameworks or libraries) (https://href.sbs)

- shunyaek.se (trying to build my own brand and identity and building solutions and products under this brand) (focus has been on the brand's design and flair, building a brand-story and setting-up a business)

Currently Working On:

- Started exploring the new world of LLMs and trying to build a chat app using which a user can talk to their documents. The intended purpose for this app is to have the user chat with financial markets docs and conduct research. Technologies: LangChain, Llama 2, FastAPI, PyPDF

My GitHub: https://github.com/HarshitDoshi and https://github.com/shunyaek

E-Mail: harshit at shunyaek dot se

Genuinely curious, what stack would you recommend for building such an analytics tool?

I don't have much to say, but I love msgspec and just got started using it a month back. Thanks for making this incredible tool!

That is honestly not how HN works. Your mistakes could be mine and other's mistakes too and we too would have loved to learn from the discussions and updates that this post would have garnered.

Even though if we cast out the mistakes or whatever it was that you had posted out of the picture for a moment, it might just have been a good point of learning and reading for anyone.

Please do not do this going forward.

Résumé/CV: https://rxresu.me/r/ef27gw

Location: India

Remote: Yes (preferred)

Willing To Relocate: Yes, but not to Bengaluru

Technologies: Python, JavaScript/TypeScript, DevOps

E-mail: harshitdoshi719[at]gmail[dot]com

I have seen and read from @dang often throughout my many years of journey with HN, but never got to know who really is @dang.

Is their true identity a secret? If not, can someone please point me out to their real name. I am just curious about their profile and would like to have a look at their LinkedIn or Twitter if they are on there.

I do not wish to pry open anyone's anonymity and so if their identity is a secret, I am fine with that. I honestly just want to have a look at their journey which sometimes LinkedIn / Twitter / GitHub / individual blogs provide.

I am old! "Axa" is seriously everything that is wrong with the way newer generations are using the language with English alphabets! I still cannot believe people do this.

I have personally set a standard or a set of rules for writing Hindi using English alphabets myself and I have seen my friends follow the same. We do not use such shortcuts/shortforms.

Hell, even "Acha" is not the correct way. It must be "Achha" (double Hs).

Speak Hindi all day, but cannot guess a single letter in a 100 tries!

Shows how much this language's usage in writing and reading has declined.