BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//txlf2026//speaker//SSEMXA
BEGIN:VTIMEZONE
TZID:US/Central
BEGIN:STANDARD
DTSTART:20251107T000000
TZNAME:CST
TZOFFSETFROM:-0600
TZOFFSETTO:-0600
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20260308T030000
RDATE:20270314T030000
TZNAME:CDT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20261101T020000
TZNAME:CST
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
SUMMARY:Don't Mess With My Data: Local AI + MCP on Linux - Rajesh Alwar
DTSTART;TZID=US/Central:20261107T170000
DTEND;TZID=US/Central:20261107T175000
DTSTAMP:20261119T204851Z
UID:pretalx-txlf2026-HQWWGD@pretalx.com
DESCRIPTION:What if you could ask an AI "is my house good for solar panels
 ?" - and get a real answer\, without sending your address leaving your mac
 hine?\n\nIn this talk\, we'll build a fully local AI agent stack on Linux 
 that evaluates solar potential for any Texas property - no cloud APIs\, no
  OpenAI key\, no data leaving your network. We'll wire together Ollama run
 ning a local LLM\, a custom MCP (Model Context Protocol) server exposing t
 ools for geocoding and solar estimation\, the free NREL PVWatts government
  API\, and PostgreSQL for local storage. The result: an agent you can ask 
 "what would solar look like on my roof?" and get a real\, human-readable a
 nswer grounded in real data.\n\nCode: https://github.com/rajeshalwar/simpl
 elocalmcp\n\nWe'll cover:\n- Running a local LLM with Ollama on Linux\n- B
 uilding an MCP server that exposes real-world tools to any AI client\n- Tu
 rning raw government solar data into plain-English savings estimates\n- Pe
 rsisting and querying results in local PostgreSQL [Advanced]\n- Running th
 e whole stack as a systemd service - no always-on laptop required [Advance
 d]\n\nTexas gets 200+ sunny days a year and some of the most volatile elec
 tricity prices in the country thanks to ERCOT. This is a practical talk ab
 out what local\, open source AI can actually do - on your hardware\, on yo
 ur terms\, using a protocol you can reuse for your own tools tomorrow.\n\n
 As the industry shifts from expensive frontier models toward cheaper\, opt
 imized inference this talk shows what's possible when you run capable AI a
 gents directly on hardware you own.
LOCATION:Lil Tex
URL:https://pretalx.com/txlf2026/talk/HQWWGD/
END:VEVENT
END:VCALENDAR
