This text briefly introduces the content in the page.
Projects

Short Video Production Platform with GenAi

This pro­ject con­sis­ted of two inter­con­nec­ted sub­pro­jects. 1️⃣ Sto­ry Gene­ra­ti­onThe first part focu­sed on buil­ding a sys­tem capa­ble of con­s­truc­ting short nar­ra­ti­ves — ran­ging from 30 seconds to 5 minu­tes — based on a user prompt. The main chall­enge was desig­ning an agen­tic work­flow that could auto­no­mously crea­te coher­ent, enga­ging, and con­tex­tual­ly con­sis­tent sto­ries from

Projects

Style Coherent Fashion Recommender with GenAi

After recent deve­lo­p­ments in GenAI, par­ti­cu­lar­ly in Visi­on-Lan­guage Models (VLMs), the fashion pro­ject I star­ted 1.5 years ago to deve­lop an out­fit recom­men­da­ti­on sys­tem seems alre­a­dy out­da­ted. The pro­ject is out­lined here. In ear­ly 2023, using VLMs was alre­a­dy an opti­on; howe­ver, due to the insuf­fi­ci­ent matu­ri­ty of available models, I cho­se a dif­fe­rent approach

Projects

Self-Learning Facial Recognition System

This pro­ject is descri­bed in more detail below under my job expe­ri­en­ces. I’m men­tio­ning it here sepa­ra­te­ly becau­se I spent almost 2 years working on the sys­tem and deve­lo­ped deep exper­ti­se in self-lear­­ning AI archi­tec­tures, lar­­ge-sca­­le data pipe­lines, and SAP and AWS-based inte­gra­ti­ons. Unfort­u­na­te­ly, due to strict secu­ri­ty and con­fi­den­tia­li­ty agree­ments, I’m not able to

Projects

ML Based Outfit Recommender System

Fashion is high­ly sub­jec­ti­ve, making it dif­fi­cult to deve­lop a sys­tem that can uni­ver­sal­ly recom­mend sty­lish and diver­se out­fits. Indi­vi­du­al pre­fe­ren­ces, sar­to­ri­al liter­acy, and per­so­nal expe­ri­en­ces great­ly influence how peo­p­le per­cei­ve and cate­go­ri­ze styl­es. Deve­lo­ping an algo­rithm that can navi­ga­te the­se nuan­ces and deli­ver per­so­na­li­zed, accu­ra­te recom­men­da­ti­ons is a com­plex task.

Projects

Automated Image Captioning PoC

I was respon­si­ble for the enti­re tech­ni­cal con­cept, fea­si­bi­li­ty stu­dy, and cli­ent com­mu­ni­ca­ti­on. Over the cour­se of the pro­ject, I deve­lo­ped a deep learning–based sys­tem for auto­ma­ted image cap­tio­ning. The main chall­enge was to com­bi­ne mul­ti­ple data sources — inclu­ding the image its­elf — into a sin­gle gram­ma­ti­cal­ly cor­rect Ger­man sen­tence describ­ing what a per­son

Projects

Analysis of EEG data from hypnotized individuals

I con­duc­ted an in-depth inves­ti­ga­ti­on of a new­ly coll­ec­ted EEG data­set to explo­re whe­ther mea­sura­ble dif­fe­ren­ces exist bet­ween indi­vi­du­als with high and low hyp­no­tiza­bi­li­ty. The stu­dy focu­sed on two key aspects: spec­tral power and func­tion­al brain con­nec­ti­vi­ty. After an exten­si­ve lite­ra­tu­re review and the appli­ca­ti­on of advan­ced sta­tis­ti­cal methods, I found that spec­tral power alo­ne

Do you want to boost your business today?

This is your chance to invite visitors to contact you. Tell them you’ll be happy to answer all their questions as soon as possible.