𝙳𝚎𝚙𝚊𝚛𝚝𝚎𝚖𝚎𝚗 𝙿𝚎𝚗𝚎𝚛𝚊𝚗𝚐𝚊𝚗 𝚁𝙸

whatever

[𝙳𝚎𝚙𝚊𝚛𝚝𝚎𝚖𝚎𝚗 𝙿𝚎𝚗𝚎𝚛𝚊𝚗𝚐𝚊𝚗 𝚁𝙸 @zombiewarnet]

@Carlyesteh12 ini angkatan ke 5 kah?

[𝙳𝚎𝚙𝚊𝚛𝚝𝚎𝚖𝚎𝚗 𝙿𝚎𝚗𝚎𝚛𝚊𝚗𝚐𝚊𝚗 𝚁𝙸 @zombiewarnet]

@sunflower_rg @KasusBasahairma gada itu istilah PHK tapi di bekukan mas, laporin aja.

[𝙳𝚎𝚙𝚊𝚛𝚝𝚎𝚖𝚎𝚗 𝙿𝚎𝚗𝚎𝚛𝚊𝚗𝚐𝚊𝚗 𝚁𝙸 @zombiewarnet]

@myohan_shi @SeekHustle waw KLO ini bisa langsung ketauan sih dari domainnya

[𝙳𝚎𝚙𝚊𝚛𝚝𝚎𝚖𝚎𝚗 𝙿𝚎𝚗𝚎𝚛𝚊𝚗𝚐𝚊𝚗 𝚁𝙸 @zombiewarnet]

@Sipitungscbd KLO diambil cepe engeh ga ya

[𝙳𝚎𝚙𝚊𝚛𝚝𝚎𝚖𝚎𝚗 𝙿𝚎𝚗𝚎𝚛𝚊𝚗𝚐𝚊𝚗 𝚁𝙸 @zombiewarnet]

@lukasabetan Simon "gus" riley

[𝙳𝚎𝚙𝚊𝚛𝚝𝚎𝚖𝚎𝚗 𝙿𝚎𝚗𝚎𝚛𝚊𝚗𝚐𝚊𝚗 𝚁𝙸 @zombiewarnet]

@iPopBase pindah ke Reddit lah udah bosen di mastodon

[Khairallah AL-Awady @eng_khairallah1]

this is f*cking gold

Andrej Karpathy joined Anthropic five weeks ago.

A friend on his team just showed me the exact LOOPS.md file he actually uses.

I dropped it into my setup. The very first response was different.

Not slightly different. Completely different.

Claude stopped giving generic answers and started working exactly the way I think.

You don't talk to the model anymore. You build the system that talks to the model for you.

Bookmark it before it gets lost in your feed.

Read it now, then check the article below.

[𝙳𝚎𝚙𝚊𝚛𝚝𝚎𝚖𝚎𝚗 𝙿𝚎𝚗𝚎𝚛𝚊𝚗𝚐𝚊𝚗 𝚁𝙸 @zombiewarnet]

@remotesimple @deepseek_ai @Alibaba_Qwen @Zai_org yes, deepseek udah worth lah buat daily yang ga perlu banyak case rumit

[Movez @0xMovez]

Ex-Google engineer explained AI agent loops, harness, evals in 20 minutes - better than 500$ courses.

trace every run → judge it with an LLM → diagnose → fix → ship.

That loop is how agents self-improve over time.

Agent loops + memory + harness + evals - thats the stack.

Watch it, then save the framework below.

[Alejo @ecommartinez]

2. https://t.co/LcPZZPX7Qd

El scraper número uno de GitHub. Convierte cualquier web en markdown limpio y listo para LLM, más rápido que los servicios de pago. Olvídate de API keys, cuentas o pagos por página. Un desarrollador lo montó en unos días después de hartarse de un scraper de pago de 16 dólares. 51 mil estrellas. Apache 2.0.

[𝙳𝚎𝚙𝚊𝚛𝚝𝚎𝚖𝚎𝚗 𝙿𝚎𝚗𝚎𝚛𝚊𝚗𝚐𝚊𝚗 𝚁𝙸 @zombiewarnet]

@fajar17 njir jokes jaman gw sd

[oxbfnohome @OxbfnoHome]

INI SKILL BUAT AGENT LU BUAT JAGO ANALISA PART 1

kemarin ini kita udah belajar tentang beberapa command penting yang bisa di gunakan di terminal dan chat,sekarang waktunya lu ngasih skill yang cocok sesuai dengan kebutahan agent mu

bagi yang belum tau apa itu skill, skill ini layaknya kehidupan, ada beberapa orang yang bisa masak, jadi kuli bangunan, coding, analisis market ,trading keripto,itu semua adalah skill, build ai agent termasuk skill yang susah dipelajari,tidak semua orang paham gimana bikin agent sendiri, yang bisa melakukan analisa, melakukan tugas" sendiri,sampai melakukan trading

sebelum lu masukan skill lu harus membedakan agent mana yang ingin lu masukin skill,jadi penting buat lu untuk bikin beberapa agent khusus,jangan semua skill lu gabungin di satu agent

seperti contoh,gw sekarang memiliki 4 agent,agent utama buat ngatur semua agent dan analisis market , agent 2 khusus nft hunter,agent 3 untuk coding,agent 4 khusus bug bounty,semua agent skillnya berbeda beda,tidak ada yang sama, jadi lu harus bisa membedakan agent mana yang ingin tambahkan skillnya,supaya tidak terjadinya eror

ini skill buat agent lo buat jago melakukan analisa

agent-reach
https://t.co/fBAGwydb2Z
intinya agent lu bisa baca & cari konten dari 16 platform seperti youtube,github, reddit, X, Bilibili, Weibo, web biasa, semuanya gratis, tanpa bayar API

byob
https://t.co/TusEHpZsJt
skill ini sedikit ribet masangnya karna harus masukin extensions byobnya ke crome/edge kalian,gunanya untuk supaya agent mu bisa akses crome lu sendiri,jika di aktifkan nantik crome lu bisa gerak sendiri sesuai dengan perintah lu

crypto skill
https://t.co/hTxmO1MX0n
sesuai dengan namanya skill ini digunakan untuk analisa crypto market

lu tinggal kasih link yang gw bagikan ke agent lu,setelah itu agent lu akan pasang skillnya sendiri

selamat mencoba

[𝙳𝚎𝚙𝚊𝚛𝚝𝚎𝚖𝚎𝚗 𝙿𝚎𝚗𝚎𝚛𝚊𝚗𝚐𝚊𝚗 𝚁𝙸 @zombiewarnet]

@punjikun @txtdrimedia itulah hasil dari jarang membaca

[𝙳𝚎𝚙𝚊𝚛𝚝𝚎𝚖𝚎𝚗 𝙿𝚎𝚗𝚎𝚛𝚊𝚗𝚐𝚊𝚗 𝚁𝙸 @zombiewarnet]

@atokdalangs4 itu akun yg lain yg pertama komen bisa dilacak juga ga ya? ko bisa kesana njir pikirannya bedebah

[𝙳𝚎𝚙𝚊𝚛𝚝𝚎𝚖𝚎𝚗 𝙿𝚎𝚗𝚎𝚛𝚊𝚗𝚐𝚊𝚗 𝚁𝙸 @zombiewarnet]

@B3doel___ anjoyy Bekasi makin seperti Gotham City coy

[𝙳𝚎𝚙𝚊𝚛𝚝𝚎𝚖𝚎𝚗 𝙿𝚎𝚗𝚎𝚛𝚊𝚗𝚐𝚊𝚗 𝚁𝙸 @zombiewarnet]

@txtdrbekasi kaya yang ngasih gratis aja njirr. mentang2 kaga ada kompetitor

[𝙳𝚎𝚙𝚊𝚛𝚝𝚎𝚖𝚎𝚗 𝙿𝚎𝚗𝚎𝚛𝚊𝚗𝚐𝚊𝚗 𝚁𝙸 @zombiewarnet]

@pistahsyo nama course.a apa mba untuk IT yang berhubungan SAP?

[𝙳𝚎𝚙𝚊𝚛𝚝𝚎𝚖𝚎𝚗 𝙿𝚎𝚗𝚎𝚛𝚊𝚗𝚐𝚊𝚗 𝚁𝙸 @zombiewarnet]

@realmadridindo1 njir udah kaya formasi Ted Lasso yg American football

[Suraj Sharma @suraj_sharma14]

If I had 6 months to become an Agentic AI Engineer.

I'd do this.

Stage 1: Python + Async Foundations
asyncio, FastAPI, event-driven architecture, error handling, API integration patterns.

Stage 2: LLM Fundamentals for Agents
Context management, model routing, token economics, latency tradeoffs, failure modes.

Stage 3: Tool Calling + Structured Outputs
Pydantic validation, function calling schemas, error recovery, dynamic tool discovery.

Stage 4: Memory + State Management
Short-term buffers, long-term vector recall, context compression, cross-session sync.

Stage 5: Single Agent Workflows
ReAct loops, plan-and-execute, self-reflection, iteration limits, graceful degradation.

Stage 6: Multi-Agent Orchestration
LangGraph/CrewAI, supervisor patterns, message passing, conflict resolution, handoffs.

Stage 7: Human-in-the-Loop Systems
Uncertainty detection, approval gates, audit trails, resume logic, intervention points.

Stage 8: Evaluation + Quality Assurance
Automated eval harnesses, LLM-as-a-judge, regression testing, hallucination metrics.

Stage 9: Observability + Tracing
Distributed tracing (LangSmith/Arize), cost dashboards, latency monitoring, alerting.

Stage 10: Security + Guardrails
Prompt injection defense, output filtering, PII redaction, sandboxed execution, compliance.

Stage 11: Production Deployment
vLLM/SGLang, Kubernetes scaling, CI/CD for agents, canary releases, rollback strategies.

Stage 12: Open Source + Portfolio
Ship autonomous agents publicly, write architecture docs, record demos, contribute to libs.

Most people stay stuck watching tutorials.

Builders get hired.

(Bookmark it)

[waishnav @wshxnv]

yesterday, at the Codex hackathon, i ended up solving this problem by building an MCP connector app that turns ChatGPT into Codex, sort of.

1. you get to use GPT-5.5 Pro by exposing your local dev environment to ChatGPT, so you can use 5.5 pro for planning and handover to codex to execute it
2. you get to 2x your rate limits, just like last month, because apparently ChatGPT and Codex have separate rate limits

here is the quick demo