Polytechnic directory & coordinates, UBTER/IRDT portal S1; per-college department figures, UBTER institutional dataset S2. Districts shaded by polytechnic count. Base map © OpenStreetMap © CARTO; district boundaries from india-maps-data (CC BY 4.0).
UBTER engineering diploma syllabus & evaluation scheme 2026-27 S1.
UBTER engineering diploma syllabus & evaluation scheme 2026-27 S1.
The nine new emerging branches
2026-27 introduced nine "emerging" branches (codes 43–51): AI & ML, Automation & Robotics, Mechatronics, Communication & Computer Networking, Construction Automation, Cyber Forensics & Information Security, Gaming & Animation, Cloud Computing & Big Data, and Civil & Environmental Engineering.
How to read the confidence labels
Verified — subject, credits and hours taken directly from an official evaluation-scheme or detailed syllabus PDF. Partial — the signature subject(s) for that semester are confirmed, but the full per-semester list was not exhaustively extractable (some scanned IV–VI files resisted extraction).
The full course structure
Every captured subject — hours, credits, type, version. Filter, sort, search.
Every captured subject across all branches: code, theory/practical hours per week, credits, course type, the confidence label and which syllabus version it comes from. Filter by branch, semester or confidence; sort any column; or search within the table.
UBTER engineering diploma syllabus & evaluation scheme 2026-27 S1.
| Branch | Sem | Code | Course Title | Th | Pr | Cr | Type | Confidence | Version |
|---|
Unit-level topics, labs & textbooks
Unit-level topics, labs and textbooks. Pick a branch to expand its subjects.
For the subjects captured in depth, the unit-level topic breakdown, lab/practical outcomes and prescribed textbooks. Pick a branch to expand its subjects.
UBTER detailed syllabus PDFs, 2026-27 S1.
| Domain | Branch | Subject | Textbook(s) |
|---|
Compiled from the prescribed-books field of the UBTER detailed syllabus S1.
Every branch, at a glance
Coverage across all branches — version, semesters captured, depth and status.
Coverage map across all engineering branches: which syllabus version applies, how many semesters and subjects were captured, the capture depth, and the collection status with notes.
UBTER engineering diploma syllabus 2026-27 S1.
| Branch | Version | Capture status | Notes |
|---|
UBTER engineering diploma syllabus 2026-27 S1.
Issues found in the source portal — empty branches, duplicate or blank entries — worth flagging to UBTER / IRDT. Listed verbatim from the workbook.
| Code | Branch | Issue type | Detail |
|---|
Compiled from the UBTER source workbook S1.
The 70 government polytechnics
All 70 government polytechnics — district, type, coordinator, website, location.
Directory of all government polytechnics with district, type, principal/coordinator, website and location. Contact phone numbers are not included. The interactive map with click-to-load profiles is on the Overview tab.
Polytechnic directory, UBTER/IRDT portal S1; institutional dataset for department, student & faculty figures S2; intake/staff for unlisted campuses, AICTE records S3.
Provided estimate, cross-checked against UBTER JEEP counselling and college fee listings S8. Government-polytechnic diploma totals reported elsewhere cluster around ₹24,000–₹39,000 across three years depending on college, branch and category.
| # | Polytechnic | District | Type | Depts | Students | Faculty | Branches running | Website |
|---|
Branch reach across the state
The by-branch view: how many colleges offer each branch, and which ones.
The complement to the directory above (which lists branches per campus). Here the relationship is pivoted by branch: a chart of how many polytechnics offer each branch, and an index listing every college that runs it.
Course-to-college mapping, Technical Education Department / UBTER S1.
Course-to-college mapping, UBTER S1.
Course-to-college mapping, UBTER S1.
Department-wise students & faculty
Real per-branch student & faculty counts — 67 of 70 campuses.
Real per-branch student and faculty counts from the official UBTER institutional dataset, covering 67 of the 70 campuses (the other 3 have verified department lists but unpublished headcounts). Foundation units (Basic Sciences, Mathematics, English, Pharmacy, etc.) are listed separately from engineering branches.
UBTER official institutional dataset S2; department lists for 3 unmatched campuses, AICTE approval records S3.
UBTER official institutional dataset S2.
| Department | Kind | Colleges | Students | Faculty | Student : Faculty |
|---|
UBTER official institutional dataset S2.
| Polytechnic | District | Department | Students | Faculty |
|---|
UBTER official institutional dataset S2; AICTE records for unlisted campuses S3.
Placements · 2025-26
Official UKDTE placement-portal figures for the current session.
Official figures from the UKDTE Training & Placement Cell portal (placement.ukdte.in ↗), the Directorate of Technical Education's in-house portal running campus drives and Rojgar Melas for the state polytechnics. Figures below are embedded from the portal's dashboard API for the current session; use “Refresh from portal” to attempt a live update.
UKDTE Training & Placement Cell portal — dashboard API S4; company-wise placement report 2025-26 S5; placement-cell meeting notes S7.
Funnel from UKDTE portal figures S4; placement rate & gap reconciled with meeting notes S7; salary from portal branch averages S4.
UKDTE Placement Cell portal — placement by source S4.
UKDTE Placement Cell portal — placed by district S4.
UKDTE Placement Cell portal — branch-wise salary S4.
UKDTE Placement Cell portal — branch registration & gender S4.
UKDTE "Company Wise Placed Student 2025-26" report S5.
The portal restricts direct browser requests (CORS), so a live refresh from this file usually won’t go through. To update: open the portal ↗, copy the dashboard API response(s) from DevTools → Network, and paste below.
Gaps identified
A data-driven read on where the 2026-27 curriculum stands — for a polytechnic student.
Framed for what matters to a polytechnic student — employability and hands-on skill, not academic depth. Every gap shows its finding, the analysis behind it, and the exact method used to compute it, so nothing is a black box. Figures are recomputed live from the embedded syllabus and placement data each time you rebuild.
Computed from the UBTER syllabus dataset S1 and the UKDTE placement data S4S5.
Separate from the curriculum gaps below, this build has its own data-coverage limits worth stating plainly: about — of captured course rows have an unextracted credit value (shown as 0), and several branches' scanned IV–VI syllabus PDFs resisted full extraction, so per-semester capture is uneven. These are limitations of this compilation, not of the official syllabus, and they don't affect the gap findings (which use ratios and titles, not absolute credit sums). They will shrink as extraction improves.
Acronyms & abbreviations
Every acronym across the syllabus and sources, expanded in plain English.
Every acronym and abbreviation that appears across the syllabus, branch names, detailed content and source documents — expanded in plain English. Grouped by kind; use the search box to jump to one.
Expanded from UBTER syllabus & source documents S1.
Where every figure comes from
The portal pages and PDF links behind every figure in this compilation.
The portal pages and direct PDF links behind this compilation. Everything traces back to the official UBTER / IRDT Uttarakhand portal.
| Branch / Document | Sem | Syllabus set | Link |
|---|
How this dashboard refreshes — and how it was built
How the dashboard refreshes, and the journey of how it was built.
Two halves to this tab. First, the live mechanism — one click triggers the refresh pipeline and you can watch counters update in place. Then, the journey behind it: the phases of conversation between a human stakeholder and an AI build agent that shaped the artifact. For the technology stack itself, see the Tech tab.
The dataset embedded in this page can in principle be refreshed by an automated agent — the same kind of agent that built this site. The button simulates that pipeline end-to-end and applies an illustrative diff to the in-memory data, then re-renders the counters at the top. The network calls are simulated (a static HTML file cannot reach the UBTER portal); the change is in-memory only — reload to restore the baseline.
The trigger above simulates a refresh in the browser. This section is the real thing: a self-hosted n8n workflow plus a local LLM that can mutate this dashboard on your own machine — the same way the NHM reference site is designed to be edited by agents. You describe a change in plain language (“add the FY 2026-27 placement figures”, “mark GP Berinag's CSE lab as established”); the agent reads the embedded JSON data blocks, applies the edit, re-runs the build-time validation, and writes a new HTML file. Nothing leaves your hardware.
index.html and extracts the #appdata / #infradata JSON blocks.<script>).index.html and let the static server pick it up. On red, return the failing check.| Node | Type | Role |
|---|---|---|
| Change request | Webhook | Entry point — receives the plain-language edit and any structured fields. |
| Load site | Read Binary File | Reads the current index.html from the mounted volume. |
| Extract data | Code (JS) | Pulls the embedded JSON blocks out of the HTML for the agent to reason over. |
| Plan patch | AI Agent (Ollama) | Local LLM proposes a minimal JSON patch and the relevant validation scope. |
| Apply patch | Code (JS) | Merges the patch, re-injects the JSON, regenerates the HTML. |
| Validate | Execute Command | Runs the build-check script; non-zero exit blocks the write. |
| Write / respond | Write File · Respond | On pass, writes the new file; on fail, returns the failing rule to the caller. |
services:
n8n:
image: docker.n8n.io/n8nio/n8n:1.74.0 # pin the version, don't use :latest
restart: unless-stopped
ports:
- "5678:5678"
environment:
- N8N_BASIC_AUTH_ACTIVE=true
- N8N_BASIC_AUTH_USER=admin
- N8N_BASIC_AUTH_PASSWORD=change-me
- N8N_ENCRYPTION_KEY=${N8N_ENCRYPTION_KEY} # keep in .env, never commit
- WEBHOOK_URL=http://localhost:5678/
- EXECUTIONS_DATA_PRUNE=true
- EXECUTIONS_DATA_MAX_AGE=168 # prune logs after 7 days
- OLLAMA_HOST=http://ollama:11434
volumes:
- ./n8n_data:/home/node/.n8n # persists workflows + credentials
- ./site:/data/site # the dashboard HTML the agent edits
depends_on:
- ollama
ollama:
image: ollama/ollama:0.5.4 # local LLM — no data leaves the host
restart: unless-stopped
ports:
- "11434:11434"
volumes:
- ./ollama_data:/root/.ollama
web:
image: caddy:2.8-alpine # serves ./site as a static site
restart: unless-stopped
ports:
- "8080:80"
volumes:
- ./site:/usr/share/caddy:ro# 1. generate an encryption key once
echo "N8N_ENCRYPTION_KEY=$(openssl rand -hex 16)" > .env
# 2. start the stack (n8n + local LLM + static web server)
docker compose up -d
# 3. pull a local model for the agent
docker compose exec ollama ollama pull llama3.1:8b
# 4. open n8n at http://localhost:5678, import the workflow,
# drop this dashboard's index.html into ./site/ — the agent edits it in place.Pattern adapted from the NHM Uttarakhand reference site's self-host pipeline and current n8n + Docker Compose self-hosting guidance (volume-mounted data, pinned image, local Ollama inference, basic-auth + encryption key). This dashboard is a single static index.html with all data in embedded JSON, which is exactly what makes it cleanly agent-mutable: the agent edits data, not layout.
How this dashboard came to be
Built in an iterative dialogue between a human stakeholder and an AI build agent capable of running tools — file reading, code execution, web search — inside a single conversation. The agent's loop and tool use are below.
The journey · phases at a glance
Each phase moved the artifact forward by one bounded request — it grew incrementally rather than being rebuilt.
The stack underneath
The technology used today — and what a fully open-source pipeline could be.
Two layers of technology. What's used today: a small set of mapping, charting, export and typography choices that produced the file you're reading. What it could become: a full open-source agentic pipeline that re-fetches the UBTER data, validates it, regenerates the HTML and deploys — on infrastructure you own, with zero SaaS dependencies and no commercial-LLM API keys leaving your network.
A · Productionising — a FOSS agentic stack
An autonomous, scheduled pipeline of free and open-source software can re-fetch the portal data, validate it, generate the HTML and deploy the site — self-hosted, no SaaS dependencies.
B · The tools used to ship the current artifact
Plots the 70 polytechnics from their lat/long. Loaded from jsDelivr CDN with an SRI hash. OpenStreetMap tiles, no API key.
leafletjs.com ↗Course-type mix, per-semester credits and branch-reach bars on the Overview and Mapping tabs. Loaded from jsDelivr CDN.
chartjs.org ↗Powers the "Excel" download button — rebuilds the full multi-sheet workbook client-side from the embedded data, with phone numbers already masked.
sheetjs.com ↗The whole type system. Serif for headings, sans for body, mono for codes and metadata. Served from Google Fonts.
ibm.com/plex ↗All HTML, CSS and JavaScript live in one document. The only external runtime requests are the three CDN assets and the map tiles. Opens in any modern browser; can be saved offline.
zero-buildUsed at build time to read all 10 sheets of the UBTER workbook and emit the structured JSON embedded in this page — including the phone-number masking step.
openpyxl ↗FOSS pipeline tool catalogue
Each role in a future self-maintaining pipeline has open options — pick what matches your infrastructure.
Visual node-based workflow builder; HTTP, Postgres, cron and Ollama nodes. Single Docker container.
Self-hostExplicit state machines for agentic flows; best fit for the validation + critic stages. Any OpenAI-compatible endpoint.
MITOne-binary local LLM runner — Llama 3.3, Mistral, Qwen, Gemma. OpenAI-compatible API, no keys leave your network.
Local LLMFetch portal pages and pull tables/text from the evaluation-scheme PDFs in stage 4.
ScrapeRegenerated HTML committed and served by a single static web server with automatic HTTPS. No CDN lock-in.
StaticLab & equipment coverage
What the directorate's own budget case says is required, installed and still missing across every branch's labs — and what closing it costs.
Built from two Technical Education Directorate budget justifications for FY 2025-26: Standard Item 26 — Computer Hardware & Software (computing labs, common computer centre, language lab, AI-IoT labs, ERP) and Standard Item 40 — Machinery, Furnishing & Equipment (lab machinery for Civil, Electrical, Mechanical, Electronics, Automobile and Pharmacy, plus workshop/physics/chemistry). Each component carries a regulatory norm (AICTE / UBTER-IRDT / PCI), a count of labs required vs. established, the shortfall, and a three-year costed plan. Together they total ₹347.18 crore across 12 branches. The Head-40 case follows an on-site inspection by 7 IIT Delhi professors (Oct 2024). Sources: Standard Item 26 S9 and Standard Item 40 S10 budget justifications, FY 2025-26.
The two budget justifications exactly as supplied — Head 26 (Computer Hardware & Software) and Head 40 (Machinery, Furnishing & Equipment). Download gives both as separate sheets in one workbook.
Norms and counts per Standard Item 26 budget justification, FY 2025-26 S9. Provenance S — a reported figure from the directorate's budget document, point-in-time for FY 2025-26.
Cost lines and three-year phasing per Standard Item 26 S9. Figures are the department's estimates at prevailing market rates and may vary in execution.
Campus lists transcribed from Standard Item 26 S9. Names normalised to the directory where possible; a few spelling variants remain as printed in the source.
Per-item cost table, Standard Item 26 S9 — "as per prevalent market rates, may vary as per actual market condition."
Per-branch lab norms and counts, Standard Item 40 budget justification, FY 2025-26 S10. Provenance S — reported figures, point-in-time for FY 2025-26; the gap analysis was informed by an on-site IIT Delhi inspection (Oct 2024).
Branch costs (₹195 cr) plus workshop, physics and chemistry labs (₹12 cr); three-year phasing per Standard Item 40 S10.
Head 26 (₹140.18 cr) + Head 40 (₹207 cr). Combined three-year demand: ₹132.5 cr (2025-26), ₹114 cr (2026-27), ₹100.68 cr (2027-28). Sources S9 S10.
Lab structure per branch, transcribed from Standard Item 40 S10. Used next as the basis for the syllabus-practicals → lab cross-link.
Per-institution establishment cost in ₹ lakhs, Standard Item 40 S10 — "as per prevalent market rates, may vary."