Deep Learning Researcher
Take a real problem from HCF’s work to a result worth publishing. HCF has IEEE-published research and runs a paper-writing cohort.
- Ref
- J02
- Team
- Research
- Where
- Remote, anywhere in India
All roles
This is the research seat. It exists because some of what HCF runs into — low-resource languages, sparse clinical data, models that must run cheaply — are open problems, not engineering tasks.
You would work on one of them properly: read the literature, build the experiment, be honest about the result, and write it up. HCF has published through IEEE before and runs a cohort specifically on getting to publication.
What you would do
- Own a research question from literature review to written result
- Design experiments that could actually disprove your hypothesis
- Keep reproducible records — seeds, configs, checkpoints
- Draft toward submission, and take review comments seriously
What we are looking for
- You can read a paper and reimplement its core idea
- Comfortable with PyTorch and with training runs that fail
- Enough statistics to know when a result is noise
- Prepared to report a negative result rather than bury it
These are not calculator projects or website clones. You will work on RAG systems, real databases and production AI pipelines — the kind of work you can actually talk about in an interview.