diff --git a/GPTEL_SETUP.md b/GPTEL_SETUP.md
index 8adc7a4..aef3bd4 100644
--- a/GPTEL_SETUP.md
+++ b/GPTEL_SETUP.md
@@ -59,22 +59,24 @@ Your Doom Emacs now has a comprehensive gptel setup with:
 
 ## ⌨️ Keybindings Reference
 
-All gptel commands are under `SPC a` (AI prefix):
+All gptel commands are under `SPC A` (uppercase A - AI prefix):
 
 | Keybinding | Command | Description |
 |------------|---------|-------------|
-| `SPC a g` | `gptel` | Open gptel chat buffer |
-| `SPC a s` | `gptel-send` | Send prompt to LLM |
-| `SPC a m` | `gptel-menu` | Open transient menu (change model, backend, etc) |
-| `SPC a r` | `gptel-rewrite` | Rewrite selected region with AI |
-| `SPC a c` | `gptel-add` | Add region/buffer to context |
-| `SPC a f` | `gptel-add-file` | Add file to context |
-| `SPC a q` | `gptel-quick` | Quick lookup in popup (word/region) |
-| `SPC a l` | Load buffer into session | Send whole buffer as context |
-| `SPC a b` | Send whole buffer | Send entire buffer to AI |
-| `SPC a R` | Refactor region | Refactor selected code |
-| `SPC a e` | Quick explain | Explain selected code quickly |
-| `SPC a p` | Proofread text | Fix grammar/spelling in selection |
+| `SPC A g` | `gptel` | Open gptel chat buffer |
+| `SPC A s` | `gptel-send` | Send prompt to LLM |
+| `SPC A m` | `gptel-menu` | Open transient menu (change model, backend, etc) |
+| `SPC A r` | `gptel-rewrite` | Rewrite selected region with AI |
+| `SPC A c` | `gptel-add` | Add region/buffer to context |
+| `SPC A f` | `gptel-add-file` | Add file to context |
+| `SPC A q` | `gptel-quick` | Quick lookup in popup (word/region) |
+| `SPC A l` | Load buffer into session | Send whole buffer as context |
+| `SPC A b` | Send whole buffer | Send entire buffer to AI |
+| `SPC A R` | Refactor region | Refactor selected code |
+| `SPC A e` | Quick explain | Explain selected code quickly |
+| `SPC A p` | Proofread text | Fix grammar/spelling in selection |
+| `SPC A C` | Complete at point | Trigger AI code completion |
+| `SPC A a` | Accept completion | Accept displayed ghost text completion |
 
 ### In gptel buffer:
 - `C-c RET` - Send message
@@ -82,7 +84,8 @@ All gptel commands are under `SPC a` (AI prefix):
 - `C-c C-k` - Abort request
 
 ### Auto-completion (in programming modes):
-- `M-TAB` or `C-c TAB` - Trigger completion
+- `M-TAB` or `C-c TAB` - Trigger completion (`gptel-complete`)
+- `C-c C-a` - Accept completion (`gptel-accept-completion`)
 
 ---
 
@@ -90,7 +93,7 @@ All gptel commands are under `SPC a` (AI prefix):
 
 Presets let you switch AI behavior for specific tasks:
 
-1. Open gptel menu: `SPC a m` or `C-c C-c`
+1. Open gptel menu: `SPC A m` or `C-c C-c`
 2. Select "Preset" option
 3. Choose from:
    - **coder** - Expert coding assistant (uses Claude Sonnet 4.5)
@@ -99,7 +102,7 @@ Presets let you switch AI behavior for specific tasks:
    - **quick** - Quick, concise answers (uses fast Gemini Flash)
    - **docs** - Documentation writer
 
-Or use `C-u SPC a s` to adjust settings on-the-fly!
+Or use `C-u SPC A s` to adjust settings on-the-fly!
 
 ---
 
@@ -107,15 +110,15 @@ Or use `C-u SPC a s` to adjust settings on-the-fly!
 
 ### 1. Chat with AI
 ```
-SPC a g          → Opens gptel buffer
+SPC A g          → Opens gptel buffer
 Type question    → Write your prompt
-C-c RET         → Send (or SPC a s)
+C-c RET         → Send (or SPC A s)
 ```
 
 ### 2. Rewrite Code
 ```
 1. Select code region (visual mode)
-2. SPC a r
+2. SPC A r
 3. Type instruction: "Add error handling"
 4. Review changes and accept
 ```
@@ -123,23 +126,23 @@ C-c RET         → Send (or SPC a s)
 ### 3. Add Context (Multi-file awareness)
 ```
 1. Open reference file
-2. SPC a c (adds buffer to context)
-3. Open gptel: SPC a g
+2. SPC A c (adds buffer to context)
+3. Open gptel: SPC A g
 4. Now AI knows about the reference file!
 ```
 
 ### 4. Quick Explanations
 ```
 1. Hover over word or select code
-2. SPC a q
+2. SPC A q
 3. See explanation in popup!
 ```
 
 ### 5. Code Completion
 ```
 Start typing code...
-M-TAB           → Shows AI completion as ghost text
-TAB             → Accept completion
+M-TAB or SPC A C    → Request AI completion (shows as ghost text)
+C-c C-a or SPC A a  → Accept the completion
 ```
 
 ---
@@ -188,18 +191,18 @@ Want to run AI locally for privacy?
 
 1. **Save conversations**: gptel uses org-mode by default, so just save the buffer!
 
-2. **Switch models on-the-fly**: `C-u SPC a s` before sending
+2. **Switch models on-the-fly**: `C-u SPC A s` before sending
 
 3. **Context management**:
-   - Add files in Dired: Mark files → `SPC a c`
-   - Remove context: `SPC a m` → Manage context
+   - Add files in Dired: Mark files → `SPC A c`
+   - Remove context: `SPC A m` → Manage context
 
 4. **Org-mode features**:
    - Each heading can be a separate conversation branch
    - Set properties: `M-x gptel-org-set-properties`
    - Set topic: `M-x gptel-org-set-topic`
 
-5. **Custom instructions**: Use `SPC a r` on selected code and give specific instructions
+5. **Custom instructions**: Use `SPC A r` on selected code and give specific instructions
 
 6. **Programmatic use**: See helper functions at bottom of `ai.el` for examples
 
@@ -210,7 +213,7 @@ Want to run AI locally for privacy?
 1. **Secure your API key** (see section above)
 2. **Run:** `doom sync` to install packages
 3. **Restart Emacs**
-4. **Try:** `SPC a g` to start chatting!
+4. **Try:** `SPC A g` to start chatting!
 5. **Watch:** "Every LLM in Emacs, with gptel" video (17 min)
 
 ---
@@ -235,10 +238,12 @@ Then: `M-x package-install RET transient RET`
 - Ensure proper encryption password
 - Test: `(auth-source-pick-first-password :host "openrouter.ai")`
 
-### Autocomplete not showing?
-- Enable manually: `M-x gptel-autocomplete-mode`
-- Check if in prog-mode buffer
-- May need specific model configuration
+### Autocomplete not working?
+- Trigger manually: `M-x gptel-complete` or `M-TAB`
+- Accept completion: `M-x gptel-accept-completion` or `C-c C-a`
+- Note: There's no auto-mode - you trigger completions on-demand
+- Check gptel is configured and working first (`SPC A g`)
+- Adjust context size if needed (see `gptel-autocomplete-*-context-lines` variables)
 
 ---
 
diff --git a/ai.el b/ai.el
index f23104b..078624a 100644
--- a/ai.el
+++ b/ai.el
@@ -26,9 +26,9 @@
   ;; Move cursor to end of response automatically
   (add-hook 'gptel-post-response-functions 'gptel-end-of-response)
 
-  ;; Keybindings
+  ;; Keybindings - using SPC A for AI (uppercase to avoid conflicts)
   (map! :leader
-        (:prefix ("a" . "AI")
+        (:prefix ("A" . "AI/gptel")
          :desc "gptel chat" "g" #'gptel
          :desc "gptel send" "s" #'gptel-send
          :desc "gptel menu" "m" #'gptel-menu
@@ -84,7 +84,7 @@
   (setq gptel-quick-use-context t)
 
   (map! :leader
-        (:prefix "a"
+        (:prefix "A"
          :desc "Quick lookup" "q" #'gptel-quick)))
 
 ;;; ============================================================================
@@ -95,7 +95,7 @@
   :after gptel
   :config
   (map! :leader
-        (:prefix "a"
+        (:prefix "A"
          :desc "Load buffer into session" "l" #'gptel-extensions-load-buffer
          :desc "Send whole buffer" "b" #'gptel-extensions-send-whole-buffer
          :desc "Refactor region" "R" #'gptel-extensions-refactor)))
@@ -106,18 +106,24 @@
 
 (use-package! gptel-autocomplete
   :after gptel
+  :commands (gptel-complete gptel-accept-completion)
   :config
-  ;; Enable in programming modes
-  (add-hook 'prog-mode-hook #'gptel-autocomplete-mode)
+  ;; Configure context size
+  (setq gptel-autocomplete-before-context-lines 100
+        gptel-autocomplete-after-context-lines 20
+        gptel-autocomplete-temperature 0.1)
 
-  ;; Configure keybindings for autocomplete
-  (map! :map gptel-autocomplete-mode-map
-        "M-TAB" #'gptel-autocomplete-complete
-        "C-c TAB" #'gptel-autocomplete-complete)
+  ;; Keybindings for autocomplete
+  (map! :leader
+        (:prefix "A"
+         :desc "Complete at point" "C" #'gptel-complete
+         :desc "Accept completion" "a" #'gptel-accept-completion))
 
-  ;; Optional: Configure which model to use for completion
-  ;; (setq gptel-autocomplete-model 'google/gemini-2.5-flash)
-  )
+  ;; Also add convenient keybindings in prog-mode
+  (map! :map prog-mode-map
+        "M-TAB" #'gptel-complete
+        "C-c TAB" #'gptel-complete
+        "C-c C-a" #'gptel-accept-completion))
 
 ;;; ============================================================================
 ;;; Optional: Add Ollama for local models (commented out by default)
@@ -163,7 +169,7 @@
 
 ;; Add keybindings for helper functions
 (map! :leader
-      (:prefix "a"
+      (:prefix "A"
        :desc "Quick explain" "e" #'my/gptel-quick-explain-region
        :desc "Proofread text" "p" #'my/gptel-proofread-region))
 
diff --git a/development.el b/development.el
index 622f2f4..2a4bdba 100644
--- a/development.el
+++ b/development.el
@@ -70,7 +70,17 @@
 
 (use-package! nix-mode
   :mode "\\.nix\\'"
+  :hook (nix-mode . lsp-deferred)
   :config
+  ;; Configure nil LSP server
+  (after! lsp-mode
+    (add-to-list 'lsp-language-id-configuration '(nix-mode . "nix"))
+    (lsp-register-client
+     (make-lsp-client :new-connection (lsp-stdio-connection "nil")
+                      :major-modes '(nix-mode)
+                      :server-id 'nil-ls
+                      :priority 0)))
+
   ;; Function to check if we're in a nixpkgs repository
   (defun in-nixpkgs-repo-p ()
     (when-let ((root (or (locate-dominating-file default-directory ".git")
diff --git a/init.el b/init.el
index 38edaba..9ba6faa 100644
--- a/init.el
+++ b/init.el
@@ -150,7 +150,7 @@
        ;;lua               ; one-based indices? one-based indices
        markdown          ; writing docs for people to ignore
        ;;nim               ; python + lisp at the speed of c
-       nix               ; I hereby declare "nix geht mehr!"
+       (nix +lsp)        ; I hereby declare "nix geht mehr!"
        ;;ocaml             ; an objective camel
        (org +brain
             +contacts