Skip to content
SIH Buddyby Ganeev Singh
Dev

πŸ”₯ Roast My Pick Β· SIH26096

Digital Heritage Archive for Memorials, Manuscripts & Ambedkar: AI-Powered Institutional Archive and Audio-Visual Knowledge Platform

Ministry of Social Justice and Empowerment (MoSJE)

Mild28/100

Reasonable choice. The scoreboard liked it. The scoreboard is not the one asking questions on the day.

Worth considering. An exceptional public primary corpus and a tangible kiosk deliverable in one statement β€” build it so every answer is a real quoted passage with its source, because a system that paraphrases Ambedkar is worse than one that simply finds him. Roughly 40–90 teams are expected to go here.

The receipts

Every red flag on this statement, in full. These are the four places it bites.

  1. Exhibit A

    An AI research assistant answering questions about Ambedkar's constitutional ideas risks generating positions he did not hold, which for a figure of this significance is a serious failure β€” anchor every answer to a quoted passage with its citation and refuse rather than compose

  2. It gets worse

    The statement is filed under Hardware and describes kiosks and displays, so a browser-only submission risks being judged as not answering the category

  3. Still reading?

    Archive plus search is a low-ceiling shape, so without the primary-source grounding as the centrepiece this becomes another document portal

  4. And the finisher

    Manuscript OCR on old and degraded documents is much harder than on printed text, so scope which parts of the corpus you actually digitise rather than claiming the archive

The damage report

Every score this statement earned, and what each one actually costs you.

  • Feasibility

    4/5

    Actually buildable, which on this slate is rarer than it sounds. Do not squander it on scope.

    The corpus is genuinely exceptional and fully public β€” the Constituent Assembly debates are completely digitised and freely available, the collected writings and speeches are published by government, and the national digital library is linked in the statement itself, so you start with real primary sources rather than assembling them.

  • Innovation scope

    3/5

    Mildly interesting. The novelty will not carry the room; the build has to.

    The features are listed but nothing about the retrieval design, the kiosk interaction model or the timeline structure is prescribed, so the product design is genuinely yours within a familiar archive-and-search shape.

  • Clarity

    4/5

    The ask is unambiguous, which quietly removes your favourite excuse.

    The capabilities, the deployment context and the physical form as kiosks and displays are all stated, and the source collections are named, though nothing specifies the corpus scope or what the research assistant should do when it does not know.

  • Acceptance potential

    3/5

    Middle of the pack. This statement will not win the room for you β€” you will have to.

    The corpus is real and outstanding and the kiosk makes it tangible, but archive-plus-search is a structurally low-ceiling product and the field will judge it on execution β€” the differentiator has to be that every answer traces to a real primary source rather than the search being fast.

  • Effort

    Heavy

    Heavy. Somebody on this team is not sleeping in week three. Pick who, on purpose.

    Corpus ingestion and OCR, metadata cataloguing, semantic search, multilingual translation and narration, an interactive timeline, a kiosk interface and the physical setup is six components, with the audio-visual archival handling being more work than the text side.

  • Demo-ability

    Easy

    Easy to demo β€” and so is everyone else's. Working is the floor here, not the achievement.

    A visitor touching a screen, asking a question and getting the actual historical exchange read back to them is immediately understandable and genuinely moving, and the kiosk gives you a physical artifact in a software-heavy field.

The demo they will have already seen

Somewhere around 40–90 teams are heading here, and the description is doing the choosing for most of them. They will read the same brief, reach the same architecture, and build a version of the same demo you are planning. Being correct is the floor. If your five minutes could be swapped with the team before you and nobody in the room would notice, you have not picked badly β€” you have built predictably, which costs exactly the same and hurts more.

What survives

The ground worth standing on when the questions start.

  • The Constituent Assembly debates are fully digitised and public, which is an unusually rich and well-structured primary corpus β€” verbatim, dated, attributed to speakers β€” and semantic search over it is genuinely useful to researchers today
  • The kiosk gives you a physical deliverable that most software teams cannot match, and it fits the Hardware category the statement is filed under
  • Retrieving and reading the actual historical exchange is more powerful than any summary, so the honest design and the impressive demo are the same thing here

Nothing here is fatal. It is just the list of places this statement pushes back, and you now get to push there first.

The framing is a joke. The findings are not β€” they are the same analysis on the statement page, and every line above is attached to a score or a fact in the record. It is one opinion with its reasoning attached, so argue with it before you trust it.